A micro-grid adaptive peer-to-peer collaboration method, device, system and medium

Abstract

The present application relates to a kind of micro-grid adaptive point-to-point cooperation method, device, system and medium, wherein the method of the present application, when requiring each micro-grid to build optimal transaction volume model locally, comprehensively considers its internal resource constraint, backup constraint, operating cost and coupling relationship with external associated market, such as electricity trading market, carbon emission trading market, green certificate trading market, ensures that transaction decision is not only based on internal economy, but also actively responds to external policy and physical constraint such as carbon emission reduction and system backup demand.Carbon emission trading cost and green certificate trading income are included in the objective function, directly stimulating low-carbon energy production and consumption, and improving the low-carbon nature of the overall transaction.Considering the uncertainty of new energy output and modeling the related backup constraint as an opportunity constraint, the transaction scheme can better respond to renewable energy fluctuations, ensuring the safety and reliability of system operation, effectively balancing the economy, low-carbon and safety of the transaction scheme.

Description

Technical Field

[0001] This invention relates to the field of energy trading technology, and in particular to a microgrid adaptive point-to-point collaborative method, device, system and medium. Background Technology

[0002] Microgrids can achieve localized management of distributed resources and trade with the upper-level distribution network by constructing a localized energy autonomous system. However, due to the uncertainty and uneven distribution of distributed resources, bidirectional power flows occur between different microgrids and the upper-level grid, increasing the power transmission pressure on the distribution network and the system's operating costs. To address this, a peer-to-peer (P2P) trading model has emerged, using a trading price between the feed-in tariff and the time-of-use tariff to promote the local utilization of distributed resources, maximize the economic benefits of all stakeholders, and protect privacy. Existing P2P trading research is divided into two categories: intra-microgrid producer-consumer transactions and inter-microgrid transactions. The former focuses on reliability issues such as resistance to attacks and communication failures; the latter considers factors such as multi-energy coupling, distribution network operator competition, and carbon trading.

[0003] However, existing peer-to-peer (P2P) trading models still have certain shortcomings: On the one hand, existing studies on inter-microgrid P2P trading have neglected the impact of the electricity-carbon-green certificate joint market on the trading results, while existing research in the electricity market has shown that multi-market joint trading is more conducive to renewable energy consumption and produces better carbon emission reduction effects than single-market trading. On the other hand, most existing solution methods for inter-microgrid P2P trading are distributed iterative solution methods, which are very costly in terms of computation and communication. In addition, existing inter-microgrid P2P research ignores the impact of renewable energy output uncertainty in modeling, does not fully consider the coupling between reserve demand and energy trading caused by uncertainty, and lacks joint optimization of internal microgrid safety reserve decisions.

[0004] Therefore, a peer-to-peer transaction method that balances economy, security, and improves computational efficiency and privacy is needed.

[0005] It should be noted that the information disclosed in the background section above is only for understanding the background of the present invention, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The technical problem this invention aims to solve is "how to provide a microgrid peer-to-peer transaction processing method that can balance the economy and security of transaction schemes, while improving the efficiency of distributed transaction computation and protecting the privacy of the parties involved." To this end, this invention provides a microgrid adaptive distributed peer-to-peer collaborative method, apparatus, system, and storage medium that integrates multiple market constraints to solve the aforementioned technical problem.

[0007] The technical solution adopted by the present invention to solve the above-mentioned technical problems is as follows.

[0008] The present invention provides, in a first aspect, an adaptive distributed point-to-point collaborative method for microgrids that integrates multiple market constraints, comprising the following steps:

[0009] In each microgrid, based on its internal resource constraints, operating costs, and coupling relationship with external related markets, an optimal transaction volume model is constructed for each microgrid under different peer-to-peer transaction prices;

[0010] The optimal trading volume under multiple point-to-point trading prices is calculated in parallel, and functional relationship data reflecting the trading intentions of each microgrid under the trading price is generated.

[0011] Each microgrid uploads its function relationship data to the Global Market Controller (GMC).

[0012] GMC receives functional relationship data from all microgrids and builds a market clearing model based on this data. The market clearing model aims to maximize system benefits or minimize total costs, while satisfying transaction balance constraints and network security constraints.

[0013] GMC solves the market clearing model to obtain the final peer-to-peer transaction clearing result, and broadcasts the clearing result to each microgrid.

[0014] In some embodiments, internal resource constraints include at least one of the following: output constraints of diesel generators, new energy units, energy storage devices, ramping constraints, and reserve constraints; coupling relationships with external related markets include coupling relationships with the power grid's electricity trading market, carbon emission trading market, green certificate trading market, and backup ancillary service market; when constructing the optimal trading volume model, its objective function includes the costs or benefits arising from participating in the carbon emission trading market and the green certificate trading market; when constructing the optimal trading volume model, the uncertainty of new energy output is considered, and the reserve constraints and / or cybersecurity constraints related to uncertainty are modeled as joint opportunity constraints.

[0015] In some embodiments, the functional relationship data is a piecewise linear marginal utility function. The specific method for generating functional relationship data reflecting the trading intentions of each microgrid at the transaction price is as follows: piecewise linear fitting is performed on multiple point-to-point transaction prices and the calculated optimal transaction volume to obtain a piecewise linear function with the transaction price as the independent variable and the optimal transaction volume as the dependent variable.

[0016] In some embodiments, the functional relationship data is a piecewise linear marginal utility function, and the market clearing model is a mixed integer linear programming model. By introducing auxiliary integer variables, the nonlinear relationship caused by identifying effective constraints after incorporating the piecewise linear marginal utility function into the market clearing model is handled.

[0017] In some embodiments, considering network usage fees, GMC employs an iterative algorithm to solve the market clearing model. The iterative algorithm includes: solving an optimization model based on the current equivalent transaction price to determine the optimal transaction volume for each microgrid; fixing the transaction volume, solving an optimization model to determine the specific transaction volume distribution among microgrids and calculate network usage fees; updating the equivalent transaction price for each microgrid according to the network usage fees; and iterating repeatedly until convergence.

[0018] Without considering network usage fees, GMC employs a non-iterative algorithm to solve the market clearing model, including the following steps: constructing a market clearing model with the goal of maximizing system efficiency. The decision variables of this model include the peer-to-peer transaction volume of each microgrid and the transaction volume with the upper-level grid; transforming the functional relationship data uploaded by each microgrid into a mixed-integer linear programming model to represent the piecewise linear relationship between transaction volume and price; solving the mixed-integer linear programming model to obtain the final peer-to-peer transaction clearing price and the clearing transaction volume of each microgrid in one step.

[0019] In some embodiments, the iterative algorithm further includes an oscillation monitoring and correction step: when an oscillation is detected in the update of the equivalent transaction price, a binary search method is used to search for the optimal equivalent transaction price increment within the oscillation range.

[0020] A second aspect of this application also provides a local computing device for generating a microgrid marginal utility function. The device is deployed on the microgrid side and includes at least one processor and a memory. The memory stores a computer program, which, when executed by the processor, performs the following steps:

[0021] Obtain internal resource parameters, operating cost parameters, and external market price parameters of the microgrid;

[0022] Based on the parameters, an optimal trading volume model is constructed that takes into account internal resource constraints, operating costs, and coupling with external related markets.

[0023] The optimal trading volume model is solved in parallel under multiple discrete peer-to-peer trading prices, resulting in a series of optimal trading volumes;

[0024] Based on the peer-to-peer transaction price and the corresponding optimal transaction volume, functional relationship data of microgrids is generated, which reflects the transaction intentions of each microgrid at the transaction price.

[0025] Output function relationship data for uploading to the global market controller.

[0026] In some embodiments, the external market price parameters include carbon emission rights trading prices and green certificate trading prices, and the optimal trading volume model is an opportunity-constrained optimization model that takes into account the uncertainty of new energy output.

[0027] A third aspect of this application also provides a distributed resource peer-to-peer trading system, comprising:

[0028] The local computing device of this application corresponds to a microgrid;

[0029] The Global Market Controller (GMC), which communicates with all local computing devices, is configured to: receive functional relationship data uploaded by all microgrids; when considering network usage fees, the GMC uses the iterative algorithm of this application to solve the market clearing model; construct the market clearing model based on all functional relationship data, transaction balance constraints, and network security constraints; solve the market clearing model to obtain the peer-to-peer transaction clearing results; and broadcast the clearing results to the local computing devices corresponding to each microgrid.

[0030] A fourth aspect of this application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it implements the steps of the method of this application.

[0031] The present invention has the following beneficial effects:

[0032] The method of this invention requires each microgrid to comprehensively consider its internal resource constraints, such as the output, ramp-up, reserve constraints, and operating costs of diesel generators, renewable energy units, and energy storage, as well as their coupling with external markets, such as the electricity trading market, carbon emission trading market, and green certificate trading market, when constructing its local optimal trading volume model. This ensures that trading decisions are not only based on internal economics but also proactively respond to external policy and physical constraints such as carbon emission reduction and system reserve requirements. By incorporating carbon emission trading costs and green certificate trading revenue into the objective function, it directly incentivizes low-carbon energy production and consumption. Considering the uncertainty of renewable energy output and modeling relevant reserve constraints as opportunity constraints, the trading scheme can better cope with renewable energy fluctuations, ensure the safe and reliable operation of the system, and effectively balance the economics and security of the trading scheme.

[0033] Furthermore, the method of this invention adopts a hybrid architecture of "locally generated marginal utility functions combined with global centralized clearing". Each microgrid calculates the optimal trading volume under multiple prices locally in parallel and generates piecewise linear marginal utility functions, transforming the complex local optimization problem, which may contain nonlinearity and uncertainty, into a concise linear piecewise function for uploading. The Global Market Controller (GMC) only needs to construct a mixed integer linear programming (MILP) market clearing model based on these piecewise linear functions. By introducing auxiliary integer variables to handle nonlinearity, the computational complexity and solution time of global optimization are greatly reduced, avoiding the high computational and communication costs of traditional fully distributed iterative algorithms, and significantly improving the computational efficiency of distributed trading.

[0034] Furthermore, in the method of this invention, the core data within each microgrid is processed and optimized only locally. Only a piecewise linear marginal utility function reflecting its trading intentions is uploaded to the GMC; this function is essentially its supply / demand curve, containing only price-quantity pair information. This approach avoids the direct exposure of sensitive individual data, achieving effective market clearing while protecting the business secrets and privacy of market participants, thus fully protecting the privacy of each microgrid.

[0035] Furthermore, for complex scenarios considering network usage fees, the method of this invention employs an efficient iterative algorithm, GMC, to solve the problem. This algorithm addresses network constraints and cost allocation through a decomposition and coordination approach. The algorithm also includes oscillation monitoring and correction steps, ensuring convergence and robustness under complex conditions. This allows the method to adapt to a wider range of practical applications, enhancing its practicality and robustness.

[0036] In summary, the method of the present invention, through the synergistic effect of the above-mentioned technical features, can balance economy and security under complex market environments and system constraints, while also significantly improving computational efficiency and ensuring the protection of the subject's privacy.

[0037] Other beneficial effects of the present invention will be further described below. Attached Figure Description

[0038] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0039] Figure 1 A schematic diagram of a multi-microgrid adaptive fast distributed peer-to-peer transaction method;

[0040] Figure 2 The diagrams illustrate the feasible region and optimal solution of the optimization problem and the impact of network fees on the microgrid. (a) illustrates the feasible region and optimal solution of the optimization problem P1, and (b) illustrates the impact of network fees on the microgrid's P2P purchase / sale volume.

[0041] Figure 3 This paper proposes an implementation path for an adaptive, fast, distributed, peer-to-peer trading method for microgrids that considers the coupling relationship between electricity, carbon, and green certificates. Detailed Implementation

[0042] The embodiments of the present invention will be described in detail below. It should be emphasized that the following description is merely exemplary and not intended to limit the scope and application of the present invention.

[0043] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of embodiments of the present invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0044] The embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0045] A first aspect of this invention provides an adaptive distributed point-to-point collaborative method for microgrids that integrates multiple market constraints, characterized by comprising the following steps:

[0046] In each microgrid, based on its internal resource constraints, operating costs, and coupling relationship with external related markets, an optimal transaction volume model is constructed for each microgrid under different peer-to-peer transaction prices;

[0047] The optimal trading volume under multiple point-to-point trading prices is calculated in parallel to generate functional relationship data reflecting the trading intentions of each microgrid. In some embodiments, the functional relationship data can be a piecewise linear marginal utility function.

[0048] Each microgrid uploads its piecewise linear marginal utility function to the Global Market Controller (GMC).

[0049] GMC receives the piecewise linear marginal utility functions of all microgrids and constructs a market clearing model based on them. The market clearing model aims to maximize system benefits or minimize total costs and satisfies transaction balance constraints and network security constraints.

[0050] GMC solves the market clearing model to obtain the final peer-to-peer transaction clearing result, and broadcasts the clearing result to each microgrid.

[0051] In some embodiments, internal resource constraints include at least one of the following: output constraints of diesel generators, new energy units, energy storage devices, ramping constraints, and standby constraints.

[0052] In some embodiments, the coupling with external associated markets includes coupling with at least two of the following markets: the power grid's electricity trading market, carbon emission trading market, green certificate trading market, and backup ancillary services market.

[0053] In some embodiments, when constructing an optimal trading volume model, its objective function includes the costs or benefits arising from participation in the carbon emissions trading market and the green certificate trading market.

[0054] In some embodiments, when constructing the optimal trading volume model, the uncertainty of new energy output is considered, and the backup constraints and / or cybersecurity constraints related to the uncertainty are modeled as opportunity constraints.

[0055] In some embodiments, generating a piecewise linear marginal utility function specifically involves: performing piecewise linear fitting on multiple point-to-point transaction prices and the calculated optimal transaction volume to obtain a piecewise linear function with transaction price as the independent variable and optimal transaction volume as the dependent variable.

[0056] In some embodiments, the market clearing model is a mixed-integer linear programming model, which introduces auxiliary integer variables to handle the nonlinear relationship of the piecewise linear marginal utility function.

[0057] In some embodiments, considering network usage fees, GMC employs an iterative algorithm to solve the market clearing model. The iterative algorithm includes: solving a first optimization model based on the current equivalent transaction price to obtain the optimal transaction volume for each microgrid; fixing the transaction volume, solving a second optimization model to determine the specific transaction volume distribution among microgrids and calculating the network usage fee; updating the equivalent transaction price for each microgrid according to the network usage fee; and iterating cyclically until convergence.

[0058] In some embodiments, the iterative algorithm further includes an oscillation monitoring and correction step: when an oscillation is detected in the update of the equivalent transaction price, a binary search method is used to search for the optimal equivalent transaction price increment within the oscillation range.

[0059] Through the above design, the method of this invention requires each microgrid to comprehensively consider its internal resource constraints, such as the output, ramp-up, reserve constraints, and operating costs of diesel generators, renewable energy units, and energy storage, as well as their coupling with external markets, such as the electricity trading market, carbon emission trading market, and green certificate trading market, when constructing its local optimal trading volume model. This ensures that trading decisions are not only based on internal economics but also proactively respond to external policy and physical constraints such as carbon emission reduction and system reserve requirements. By incorporating carbon emission trading costs and green certificate trading revenue into the objective function, it directly incentivizes low-carbon energy production and consumption. Considering the uncertainty of renewable energy output and modeling relevant reserve constraints as opportunity constraints, the trading scheme can better cope with renewable energy fluctuations, ensure the safe and reliable operation of the system, and effectively balance the economics and security of the trading scheme.

[0060] Furthermore, the method of this invention adopts a hybrid architecture of "locally generated marginal utility functions combined with global centralized clearing". Each microgrid calculates the optimal trading volume under multiple prices locally in parallel and generates piecewise linear marginal utility functions, transforming the complex local optimization problem, which may contain nonlinearity and uncertainty, into a concise linear piecewise function for uploading. The Global Market Controller (GMC) only needs to construct a mixed integer linear programming (MILP) market clearing model based on these piecewise linear functions. By introducing auxiliary integer variables to handle nonlinearity, the computational complexity and solution time of global optimization are greatly reduced, avoiding the high computational and communication costs of traditional fully distributed iterative algorithms, and significantly improving the computational efficiency of distributed trading.

[0061] Furthermore, in the method of this invention, the core data within each microgrid is processed and optimized only locally. Only a piecewise linear marginal utility function reflecting its trading intentions is uploaded to the GMC; this function is essentially its supply / demand curve, containing only price-quantity pair information. This approach avoids the direct exposure of sensitive individual data, achieving effective market clearing while protecting the business secrets and privacy of market participants, thus fully protecting the privacy of each microgrid.

[0062] Furthermore, for complex scenarios considering network usage fees, the method of this invention employs an efficient iterative algorithm, GMC, to solve the problem. This algorithm addresses network constraints and cost allocation through a decomposition and coordination approach. The algorithm also includes oscillation monitoring and correction steps, ensuring convergence and robustness under complex conditions. This allows the method to adapt to a wider range of practical applications, enhancing its practicality and robustness.

[0063] In summary, the method of the present invention, through the synergistic effect of the above-mentioned technical features, can balance economy and security under complex market environments and system constraints, while also significantly improving computational efficiency and ensuring the protection of the subject's privacy.

[0064] The following further describes specific embodiments of the present invention and examples of its algorithm implementation.

[0065] Specifically, this invention is an adaptive, multi-market-constraint-integrated method for distributed peer-to-peer trading in microgrids, considering the coupling costs of the electricity-carbon-green certificate market and the coupling relationship between energy and reserves. By coupling the peer-to-peer trading model with carbon-green certificate market revenue, it improves the operational economics of multi-microgrid systems. Simultaneously, it addresses the impact of microgrids' internal safety reserve needs for dealing with uncertainties in renewable energy output on peer-to-peer energy trading, thereby enhancing the operational security of the microgrid itself during peer-to-peer trading. Furthermore, through an adaptive fast clearing method, it can adapt to both market rules with and without network usage fees, achieving non-iterative clearing in the absence of network usage fees and rapid distributed iterative clearing when network usage fees are considered. This significantly improves the distributed computing speed of peer-to-peer trading, meets real-time trading and operational requirements, and effectively protects the information privacy of each microgrid entity.

[0066] The method proposed in this invention can be summarized as follows: Figure 1 First, each microgrid executes its own security-constrained microgrid economic low-carbon energy management model locally. Under different discrete P2P trading prices, it determines the optimal scheduling decision for internal resources and the expected energy trading decision with the upper-level grid and other microgrids, and generates its piecewise linear marginal utility function locally in parallel. Then, each microgrid uploads its piecewise marginal utility function to the Global microgrid controller (GMC). The GMC, based on current market rules and distribution network parameters, uses the proposed adaptive fast distributed method to perform point-to-point market clearing and broadcasts the clearing results to each microgrid entity. Finally, each microgrid performs power exchange at its corresponding distribution network node based on the clearing results. The method proposed in this invention involves only two communications: uploading the piecewise linear marginal utility function and broadcasting the point-to-point market clearing. Therefore, microgrids belonging to different operators do not disclose internal power generation resources, network topology, or other information when uploading their piecewise linear marginal utility functions, enhancing the privacy of distributed computing while significantly reducing communication requirements and costs. In addition, each microgrid trades carbon allowances and green certificates with a Carbon Allowance and Green Certificate Balancer (CaGcB), and then the CaGcB and the National Green Certificate and Renewable Energy Trading Center jointly calculate the carbon-green certificate market cost of the entire microgrid system.

[0067] Specific embodiments of the present invention include the following steps:

[0068] Step 1.1 Microgrid Economic Low-Carbon Energy Management Model with Security Constraints: Jointly optimize the optimal resource decision within the microgrid and the expected energy trading volume with the upper-level distribution network and other microgrids.

[0069] First, this invention establishes a safety-constrained microgrid economic low-carbon energy management model that considers energy reserve coupling relationships and the costs of electricity-carbon-green certificate coupled trading. This model helps microgrid operators achieve optimal output and reserve capacity configuration schemes for distributed diesel engines, distributed renewable energy sources, and energy storage devices within the microgrid, and determine optimal electricity-carbon-green certificate trading decisions with the upper-level distribution network and other microgrids under a given P2P trading price. Simultaneously, considering line power flow and node voltage constraints at the distribution network level, it ensures that the P2P trading volume between microgrids does not cause operational risks to the distribution network.

[0070] Step 1.1.1 Objective function of microgrid economy low-carbon energy management model.

[0071] The objective function of the proposed microgrid economic low-carbon energy management model is shown in equation (1), which includes the operating cost of each microgrid. The cost of electricity transactions with the upper-level power grid or other microgrids The cost of carbon allowance trading with carbon-green certificate balancers Transaction costs of green certificates Ancillary service costs in transactions with the backup ancillary services market And network usage fees generated during P2P transactions. As shown in equations (2a)-(2f), the operating costs of a microgrid include the fuel cost and up / down reserve cost of the distributed diesel generator's base output, and the charging and discharging loss cost and up / down reserve cost of the energy storage device. Furthermore, since each microgrid trades at a P2P price between the grid connection price and the electricity price, both the buyer and seller microgrids can profit, thus both participating microgrids share the network usage fee equally.

[0072] (1)

[0073] (2a)

[0074] (2b)

[0075] (2c)

[0076] (2d)

[0077] (2e)

[0078] (2f)

[0079] Where: set They represent microgrids respectively. mA collection of distributed diesel generators and energy storage devices. N This indicates the integration of microgrids. Indicates the running time period. These represent diesel generators. g exist t The baseline output at any given time, upward reserve, and downward reserve. and These represent diesel generators. g The fuel cost function and reserve cost coefficient are determined, where the fuel cost function can be a piecewise linear cost function as used in existing technologies. Indicates energy storage device e exist t The charging and discharging losses at any time, upward reserve and downward reserve. This represents the corresponding charging and discharging loss cost coefficient and reserve cost coefficient. They represent t Moment Micro Network m From micronet n Electricity purchased and sold at the point of sale They represent microgrids respectively. m exist t Electricity purchased and sold from the upper-level power grid at all times. These represent the time-of-use electricity price, the grid connection price, and the P2P transaction price, respectively. Let represent the carbon allowances purchased by microgrid m from CaGcB at time t, the carbon allowances sold, the green certificates purchased, and the green certificates sold, respectively. These represent the carbon quota trading price and the green certificate trading price, respectively. These represent upward and downward reserves purchased from the ancillary services market, respectively. This indicates the price of purchasing spare parts from the ancillary services market. microgrid m Hewei.com n The network usage fee between them.

[0080] Step 1.1.2 Constraints of the microgrid economy low-carbon energy management model.

[0081] To effectively manage resources within a microgrid and ensure its operational safety, the constraints of the microgrid economic low-carbon energy management model include active power balance constraints based on predictive information, baseline power limits to ensure the availability of various resources, carbon quotas and green certificates, reserve constraints to address / consider uncertainties, and network constraints to ensure power flow and voltage security. The first three are modeled as deterministic constraints, while the latter two are modeled as joint opportunity constraints. The various constraints are described in detail below.

[0082] Constraint (3) indicates that the active power balance of the microgrid system is achieved by scheduling internal power generation resources and trading electricity with the external power grid. Constraints (4a)-(4d) respectively represent the capacity limits of the reference active power output and reserve of the diesel generator set, the capacity limit of reactive power output, the ramp rate limit of reserve delivery within the time period, and the ramp rate limit between the reference power and the output at the initial moment. Uncertain renewable energy output is represented as the superposition of the predicted value and the prediction error, as shown in equation (5a), where the prediction error is assumed to follow a mixed Gaussian distribution, and the corresponding parameters can be obtained from historical data. Constraint (5b) represents the reactive power output limit of new energy. Constraints (6a)-(6c) respectively represent the capacity limit constraints of the reference power and reserve of energy storage, the energy state constraint, and the charging and discharging loss constraint.

[0083] (3)

[0084] (4a)

[0085] (4b)

[0086] (4c)

[0087] (4d)

[0088] (5a)

[0089] (5b)

[0090] (6a)

[0091] (6b)

[0092] (6c)

[0093] In the formula: They represent microgrids respectively. m The collection of internal new energy generating units and busbars, and They represent microgrids respectively. m Internal new energy units k exist t The predicted active power output, random prediction error, actual active power output, and reactive power output at any given time. microgrid m Internal new energy units k The reactive power output capacity limitation They represent microgrids respectively. m Internal diesel generator set gMaximum / minimum active power output boundary, reactive power output limit, uphill rate limit, downhill rate limit. Indicates diesel generator set g The effort exerted at the initial moment, They represent microgrids respectively. m Internal energy storage device e exist t The active power output at each moment and the energy value at the initial moment. They represent energy storage devices. e The upper limit of charge / discharge power and the upper / lower limit of energy. These represent the charging efficiency and discharging efficiency of energy storage, respectively. microgrid m internal nodes i In t Active load at any given time This refers to the scheduling interval.

[0094] As shown in (5a), there is a random deviation between the actual output of new energy and the predicted value. To ensure that the system has sufficient flexibility to adjust and control the output error and avoid power imbalance, the reserve constraints of diesel generator sets and energy storage are modeled as (7a) and (7b) based on affine strategy and uncertainty prediction error. The affine strategy refers to allocating the unbalanced power caused by uncertainty prediction error to each generator set according to the secondary frequency regulation participation factor of each generator set, and reserving corresponding reserve power for each generator set, thereby ensuring that each generator set has sufficient adjustment space during real-time secondary frequency regulation. In addition, considering that the resource characteristics of different microgrids are different, in order to avoid some microgrids lacking or having insufficient adjustment resources, this invention also considers auxiliary reserves purchased from the ancillary service market.

[0095] (7a)

[0096] (7b)

[0097] Constraints (8a) and (8d) represent the active and reactive power injections of each node in the microgrid, respectively. Considering the reserve purchases from the ancillary service market, constraints (8b) and (8c) represent the diesel generator output under positive and negative power disturbance conditions, respectively. Energy storage device output and the injected power at the upstream grid coupling node. Constraints (9a) and (9b) represent line power flow and node voltage magnitude, respectively. To reduce the conservatism of decision-making while ensuring safety, the backup constraint and network security constraint related to the uncertainty variables are modeled as a joint opportunity constraint (10), which indicates that the overall probability of defaulting on any constraint in constraint (10) cannot exceed the risk threshold allowed by the operator. Based on engineering experience, its value is usually 5% or 10%, meaning that the confidence level for all constraints to hold must not be lower than [a certain percentage]. By setting an allowable probability of default, overly conservative decisions can be avoided, thus improving the economy of decision-making. Furthermore, joint opportunity constraints, by limiting the joint probability of default of all constraints, further enhance the security of decisions, thereby facilitating the formulation of decisions that balance economy and security.

[0098] (8a)

[0099] (8b)

[0100] (8c)

[0101] (8d)

[0102] (9a)

[0103] (9b)

[0104] (10)

[0105] In the formula: They represent microgrids respectively. m Middle node i In t The active power injection, reactive power injection, and output of the reactive power compensation device at any given time. Indicates diesel generator set g New energy units k Energy storage devices e The power transfer factor to node i Represents a node i Power injection at the coupling point of the upstream power grid caused by auxiliary backup. Represents a node i Load shedding at the location, They represent the lines respectively. ij Resistance, reactance, and line capacity Represents a node i Voltage assistance, maximum / minimum boundaries, and standard voltage amplitude at the location. microgrid m The collection of internal lines, This indicates the probability of default allowed by the microgrid operator.

[0106] Furthermore, since joint chance constraints are nonlinear constraints and difficult to solve directly, this invention employs an improved sample averaging approximation method to transform the nonlinear joint chance constraints into a linear model. The specific implementation method is as follows: joint chance constraints in standard form... Let's take an example to illustrate, where Represents random variables, For random variables Distribution Inequalities containing random variables Let (11a) and (11b) be the confidence level of the joint chance constraint. Based on the sample averaging approximation method in existing technical solutions, this joint chance constraint can be transformed into a mixed-integer linear programming model based on the equivalent probability distribution of the sampled scenario, as shown in Equation (11). Constraints (11a) and (11b) are used to determine the scenario. s Under the constraints Does it hold true? If so, then the scenario indicator variable... ,otherwise This indicates a constraint violation in this scenario. Constraint (11c) is used to ensure that the probability of a constraint violation occurring does not exceed [a certain value]. The above transformation can ensure At confidence level It is true, but due to the scene indicator variable Since the variables are integers (0 and 1), the model transformed by the traditional sample averaging approximation method is still a non-convex model. Therefore, this invention employs an improved sample averaging approximation method, relaxing the scene indicator variables into continuous variables and introducing the inequality shown in equation (12), thereby transforming the nonlinear joint chance constraint into a linear model (13). It is worth noting that the scene indicator variables in equation (13c) are continuous variables between 0 and 1, rather than the integer variables (0 and 1) in the original equation (11c), and the linear model (13) also includes inequality constraints (13d). Based on this transformation method, the joint chance constraint (10) containing multiple sets of constraints can be transformed into a linear model.

[0107] (11a)

[0108] (11b)

[0109] (11c)

[0110] (12)

[0111] (13a)

[0112] (13b)

[0113] (13c)

[0114] (13d)

[0115] In the formula: for In sampling scenarios s The value below, , For non-negative values, y For non-negative auxiliary variables, [ n [] represents a set of sampling scenarios. Representing a scene s The probability of occurrence and satisfying , Indicates by and A set of sampled scenarios with rearranged relationships.

[0116] Constraints (14) and (16) are the carbon quota constraint and green certificate constraint within the microgrid system, respectively. These represent the free carbon emission allowance benchmark and carbon emission factor for diesel fuel units, respectively. This indicates that the carbon emission reduction amount behind the green certificates corresponds to the weighting required for the consumption of new energy sources. This indicates the required proportion of new energy consumption. This indicates the generation capacity of conventional fuel units and renewable energy units per MWh. k The difference in carbon emissions between them.

[0117] (14)

[0118] (15)

[0119] (16)

[0120] Step 1.1.3 Distribution network constraints to ensure that P2P transaction volume does not increase the risk of distribution network operation.

[0121] When conducting P2P transactions, in addition to meeting the security constraints of the microgrid's own system, the power injection caused by P2P transactions also needs to meet the network security constraints at the distribution network level. At this time, each node in the distribution network... t Active power injected at all times and reactive power injection Power flow of the line and the trend of no effort Node voltage amplitude As shown in equations (17a)-(17c) respectively, the network security constraints at the distribution network level can be expressed as constraint (18).

[0122] (17a)

[0123] (17b)

[0124] (17c)

[0125] (18)

[0126] In the formula: This represents the set of microgrids connected to distribution network node i. These represent the set of lines and the set of nodes in the distribution network, respectively. Indicates distribution network node j Reactive power compensation t Constant effort Indicates distribution network line ij Resistance and line transmission capacity, These represent nodes in the distribution network. j The upper and lower limits of the voltage amplitude.

[0127] Step 1.2 Adaptive Fast Distributed Peer-to-Peer Transaction Methods: including non-iterative clearing methods without network usage fees and fast iterative solution methods that consider network usage fees.

[0128] To improve the computational efficiency of P2P transactions between microgrids and protect the information privacy of microgrid entities, this section first analyzes the peer-to-peer transaction willingness of microgrids under different P2P market prices and constructs a piecewise marginal utility function that reflects the optimal transaction willingness of microgrids. Then, based on the piecewise marginal utility function, a non-iterative clearing method for network usage fees and a fast iterative solution method considering network usage fees are proposed.

[0129] Step 1.2.1 Marginal utility function analysis and modeling of microgrid system.

[0130] This invention defines the response relationship between microgrid P2P transaction power and P2P transaction price as the marginal utility function of the microgrid entity, which expresses the microgrid operator's willingness to participate in market transactions after taking into account its own operating conditions. Therefore, by combining the marginal utility functions of each microgrid, the P2P market between microgrids can be cleared. Thus, this section first analyzes and establishes the marginal utility function of the microgrid.

[0131] Because the microgrid economic low-carbon energy management model constructed in step 1.1, which considers the market cost coupling of electricity-carbon-green certificates and energy-reserve coupling, involves complex coupling relationships such as physical constraints and power flow security constraints of various generator sets and energy storage devices, it is difficult to directly obtain the marginal utility function of the microgrid entity with respect to P2P transaction prices. Therefore, this invention constructs the marginal utility of the microgrid through a data-driven (discrete point fitting) approach, which also conforms to the actual engineering situation. The specific operation is as follows: First, a set of discrete P2P transaction prices is generated based on the grid-connected electricity price and the time-of-use electricity price. Then, each microgrid operator solves the microgrid economic low-carbon energy management model under each discrete P2P transaction price in parallel locally, obtaining the optimal expected transaction volume of the microgrid entity under different P2P transaction prices, forming a mapping data pair between the microgrid's expected P2P transaction volume and the P2P market price. Finally, a piecewise linear fitting method is used to construct the piecewise linear marginal utility function of the microgrid in a data-driven manner. The obtained piecewise marginal utility function is expressed as equation (19), where This indicates the transaction price in the P2P market. For micro-network i In transaction price The optimal expected P2P transaction volume is given, with positive values ​​representing the purchase of electricity from the P2P market and negative values ​​representing the sale of electricity. For the piecewise linear marginal utility function, the corresponding first... k The coefficients of the segmental linear approximation It is a set of segments.

[0132] (19)

[0133] Step 1.2.2 Two-round communication non-iterative calculation clearing method without network usage fees.

[0134] Theoretically, solving the marginal utility function of each microgrid simultaneously to calculate the common feasible solution yields the clearing price of the P2P market. However, due to factors such as the trading intentions of different microgrid entities and network constraints at the distribution network level, the expected transaction value for each microgrid to maximize its own benefits cannot always be fully satisfied. Therefore, the P2P market cannot be cleared by directly solving the joint equations. This invention designs an adaptive clearing method that considers the network security constraints of the distribution network. This method is applicable to P2P market clearing under different market rules, specifically including: a two-round communication non-iterative calculation clearing method that does not consider the network usage time between P2P entities (as described in this step), and a two-round communication fast iterative calculation clearing method that considers the network usage time (as described in step 1.2.3).

[0135] First, construct the optimization problem P1 shown in equation (20), the objective (20a) of which is to maximize the revenue obtained through P2P transactions (i.e., the difference between the cost of energy trading with the upper grid and the cost of P2P market transactions) and minimize the amount of electricity purchased from the upper grid, thereby forcing each microgrid to purchase electricity from the P2P market as much as possible.

[0136] (20a)

[0137] (20b)

[0138] (20c)

[0139] (20d)

[0140] (20e)

[0141] Constraint (20b) ensures a balance in transaction power in the P2P market; constraint (20c) limits the amount of P2P transactions cleared by microgrids. It cannot exceed the optimal expected value under its marginal utility function. Constraint (20d) indicates that when the P2P transaction volume cannot meet the microgrid's surplus power transmission or insufficient power demand, the microgrid will engage in power trading with the upstream power grid. This represents the amount of power exchanged with the upstream power grid; a positive value indicates electricity purchase, and a negative value indicates electricity sale. Let represent the sets of the marginal utility function segments for purchasing electricity and selling electricity, respectively. These represent the linear coefficients of the last segment of the electricity purchase interval. Then, it represents the linear coefficient of the initial segmentation of the electricity sales area; constraint (20e) means that when each microgrid conducts P2P power trading or conducts power trading with the upper-level grid, the power injected into the distribution network will not cause the power flow and voltage limit of the distribution network. Among them, the power flow and line constraints of the distribution network have been established in (17)-(18). The active power injected into the node consists of the point-to-point trading power of the microgrid and the trading power between the microgrid and the upper-level grid. Moreover, the constraints (17)-(18) adopt the linearized power flow formula, so the abstract mathematical form is used here. To simplify the expression of the impact of the point-to-point trading power of the microgrid and the trading power between the microgrid and the upper-level power grid on the distribution network power flow in problem P1, thus avoiding cumbersome formulas. Furthermore, it should be noted that this invention embeds the power flow security constraints on the distribution network side as hard constraints into the optimization problem P1, therefore the obtained P2P clearing results can ensure the operational safety of the distribution network power flow.

[0142] Due to microgrids iOptimal expected transaction volume under the marginal utility function It's about price. The piecewise linear function should satisfy the condition constraints shown in equation (21), where This represents the inflection point of the piecewise marginal utility function of microgrid i with respect to the price of selling or buying electricity. Let the piecewise marginal utility functions represent the first and second halves of the electricity purchase and sales intervals, respectively. k Price segment boundaries, Indicates the first [unit] within the electricity purchase range k Piecewise linear coefficients, Indicates the first electricity sales zone k Piecewise linear coefficients. However, constraint (21) cannot be solved directly. Therefore, this invention introduces 01 auxiliary variables. The Big M method is used to transform the conditional constraints (21) into tractable mixed-integer linear programming (MILP) constraints (22)-(24), where Indicates the electricity purchase range number k The corresponding constraints of a segment are effective; otherwise, they are invalid constraints. Constraint (24) restricts the total quantity of sales and purchases to always exist, and only one segment of the marginal utility function is effective. Therefore, by solving the MILP problem containing a small number of 0 and 1 variables in equations (20), (22)-(24), the P2P market can be cleared in a non-iterative manner.

[0143] (twenty one)

[0144] (twenty two)

[0145] (twenty three)

[0146] (twenty four)

[0147] Step 1.2.3 considers a two-round communication fast iterative calculation clearing method for network usage fees.

[0148] While the method in step 1.2.2 can quickly clear P2P transactions between microgrids in a non-iterative manner, it fails to account for the impact of network usage fees on the transaction scheme when distribution network operators collect them. However, network usage fees affect P2P clearing power by influencing the microgrid's electricity purchase cost and sales revenue through P2P transactions, such as... Figure 2 As shown, where Figure 2 (a) in the figure shows that modifying the P2P transaction price in optimization problem P1 may cause the optimal solution of the problem to change; Figure 2(b) quantifies the reactions of buyers and sellers when network usage time is considered compared to ignoring it. Due to the increase in effective costs, the purchase volume of microgrids will decrease, and due to the decrease in profit margins, the sales volume of microgrids will also decrease.

[0149] (25)

[0150] To address the impact of network usage fees on P2P transaction schemes, this invention proposes an anti-shock iterative clearing scheme based on the equivalent cost increment (ECI) caused by network usage fees, as shown in Table 1. First, in step 1, the parameters of the iterative calculation method are initialized, including the network parameters of the input distribution network, setting the iteration number u=0, and setting the initial ECI for each microgrid. and iteration gaps Convergence threshold The relevant parameters can be set according to project requirements. In step 2, the while loop iteration begins. In step 3, the iteration count is updated, i.e., the current iteration count is... u =u+ 1. In step 4, solve optimization problem P1 to obtain the optimal total transaction volume for each micronet when there is no network usage fee. Then, fix the current total amount of P2P transactions. In step 5, the optimal energy allocation between micronets is obtained by solving the optimization problem P2 in equation (25) with the goal of minimizing network usage fees. That is, by fixing the total amount of P2P transactions in each micronet and solving problem P2, the specific P2P transaction volume between each micronet can be determined, such as the P2P transaction volume between micronet i and micronet j. Then, in step 6, the ECI of the total transaction volume for each micronet is calculated. A positive ECI represents increased purchase costs, while a negative ECI represents decreased sales revenue. Finally, the P2P transaction price for each micronet is adjusted to the P2P market price. ECI obtained from their respective previous iterations The sum of, i.e. Then, under this modified price, the modified optimization problem P1 is solved to obtain the new total P2P transaction volume. This logic is followed iteratively until the ECI obtained from adjacent iterations calculated in step 9 satisfies the convergence gap, at which point the iteration is terminated, and the final specific P2P power trading scheme between microgrids is output.

[0151] Furthermore, to avoid oscillations during ECI price increment updates, oscillation monitoring and correction mechanisms were added in steps 7 and 8, respectively. Specifically, in step 7, oscillations are detected by identifying recurring ECI values ​​in the monitoring window. If recurring ECI values ​​appear regularly, it indicates that the true equivalent ECI lies within the range of those oscillation values. The maximum length of the monitoring window can be set according to project requirements. This refers to the length of the current monitoring window, which is defined in the for loop in step 7, using monitoring window lengths of 2, 3, 4, ... to... Each window is monitored once. This invention defines an oscillation as occurring when the ECI values ​​in two windows are identical. This is because if the ECI values ​​in two adjacent monitoring windows are not identical, problem P2 will always have a new allocation result. Only when problem P2 cannot balance the optimal total transaction volume and the equivalent ECI will the same sequence of ECI values ​​repeat. If an oscillation is detected in any monitoring window, an oscillation flag is set. and initialize the search counter. k i Then, in step 8, the first if statement is executed; if it exists... This indicates the presence of oscillations, requiring an optimal ECI search. Then, the second if statement is executed; if... k i = When microgrid i first begins its binary search (i.e., at step 1), it is necessary to record the ECI interval where oscillations occur (i.e., the left and right endpoints of the initial binary search), and set the current... The left endpoint of the search Recorded as the smallest ECI value within the monitoring window where oscillations occur. , will the current The right endpoint of the search Recorded as the largest ECI value within the monitoring window where the oscillation occurred. and calculate the current Midpoint of the search The value is the midpoint between the left and right endpoints, and the first... u The iterative ECI value is set to the current Midpoint of the search Then update the search counter. The number of searches. Until the next u iteration (since oscillation has already occurred and the first binary search replacement has been performed, the number of searches at this time and thereafter is...). If the value is greater than 1, then proceed with the subsequent binary search: First, determine whether the calculated ECI of the current iteration is closer to the left endpoint or the right endpoint of the previous search interval. If it is closer to the left endpoint (i.e., the value is greater than 1), then proceed with the binary search. If the left endpoint of the current search is retained as the left endpoint of the previous search, then the right endpoint of the local search will be taken as the midpoint of the previous search. Otherwise, the left endpoint of the local search is taken as the midpoint of the previous search, and the right endpoint of the current search is retained as the right endpoint of the previous search. Then, the midpoint of the current search interval is calculated and its magnitude is assigned to the ECI of the current iteration to update the ECI input of optimization problem P1, i.e. The search indicator is updated accordingly. Following this logic, even after oscillations occur, the optimal ECI value can still be approximated through a binary search. Finally, in step 9, when the ECI of adjacent iterations meets the convergence gap, the while loop ends in step 10, and in step 11, the cleared P2P price and the specific transaction volume between each micronetwork are output. The proposed fast iterative clearing calculation method can converge to the optimal or near-optimal ECI, significantly improving the robustness and noise resistance of the solution method.

[0152] Table 1 Adaptive Fast P2P Clearing Method

[0153]

[0154]

[0155] The method of the present invention is summarized below.

[0156] like Figure 3The overall implementation path of the proposed method is summarized as follows: First, in step 1.1, a microgrid economic low-carbon energy management model considering the coupling of electricity-carbon-green certificate market costs and energy-reserve coupling is constructed. This model reflects the coupling relationship between carbon-green certificates and electricity through carbon quota constraints and green certificate trading constraints. Based on an affine strategy, the reserve demand within the microgrid to cope with uncertainty correction control is modeled, and uncertainty-related reserve constraints and network security constraints are modeled as joint opportunity constraints that balance economics and security. Then, as in step 1.2.1, each microgrid calculates the optimal expected trading volume of the microgrid economic low-carbon energy management model locally and in parallel under discrete prices, forming data pairs of (P2P trading price, expected trading volume), and fitting them to generate a piecewise marginal utility function. Subsequently, each microgrid uploads its segmented marginal utility function to the global microgrid controller, which adaptively determines the solution method based on market rules: if network usage fees are not considered, the two-round communication non-iterative calculation clearing method in step 1.2.2 is executed; if network usage fees are considered, the two-round communication anti-oscillation fast iterative calculation clearing method in step 1.2.3 is executed to clear the P2P transaction price and the P2P transaction volume of each microgrid. The clearing result is then broadcast to each microgrid, which localizes the fixed boundary P2P transaction volume and executes the microgrid economic low-carbon energy management model to output the scheduling plan for its local units. It is evident that the proposed method balances the economic efficiency of the transaction scheme with the security against uncertainty, and improves computational efficiency and the privacy of information within the microgrid through a distributed approach. Furthermore, the advantage of this method lies in the fact that it only requires two rounds of communication (i.e., marginal utility function upload and clearing result broadcast) to achieve P2P transaction clearing among multiple microgrids. In this process, the Global Microgrid Controller (GMC) adaptively selects between non-iterative and fast iterative computation based on whether network usage fees are required to obtain the peer-to-peer transaction results. Although the GMC needs to solve the aforementioned MILP problem P1 with 0 and 1 variables during the clearing calculation, the number of 0 and 1 variables in this problem is relatively small, and existing commercial solvers such as GUROBI and CPLEX have powerful and efficient MILP solving capabilities, thus effectively supporting the implementation of this method.

[0157] The following analysis uses an example of a five-micro-network interconnection system.

[0158] This section uses a microgrid interconnection system consisting of five microgrids connected in a distribution network as an example to compare and analyze the scheduling results of the traditional model that ignores the coupling cost of electricity-carbon-green certificates and the microgrid low-carbon economic energy management model proposed in this invention, to illustrate the economic advantages of the proposed method. Simultaneously, it compares and analyzes the calculation results of the traditional Alternating Direction Method of Multipliers (ADMM), which requires multiple rounds of communication and iterative calculations, and the proposed method, to illustrate the computational advantages of the proposed method.

[0159] First, the violation probability of the joint opportunity constraint in the microgrid low-carbon economy energy relationship model is set to 0.05, and the joint opportunity constraint is solved using the improved sample averaging approximation method based on 500 sampled scenarios. The Gaussian mixture distribution parameters of the new energy prediction error are calculated using the Scikit-learn library. The charging and discharging efficiency of energy storage is set to 0.90, and the discharging efficiency is set to 0.95. The discrete price step size for calculating the marginal utility function is 0.01 yuan / kWh. Simulations are performed on an AMD Ryzen 78845H laptop with 32GB RAM using MATLAB R2020a and Gurobi 12.0.1.

[0160] The specific microgrid configurations are as follows: Microgrid 1, with 15 nodes and containing 105MW diesel generators, 35MW energy storage, and 175MW of renewable energy, is connected to distribution network node 7; Microgrid 2, with 6 nodes and containing 20MW diesel generators, 30MW energy storage, and 120MW of renewable energy, is connected to distribution network node 13; Microgrid 3, with 7 nodes and containing 70MW diesel generators, 30MW energy storage, and 50MW of renewable energy, is connected to distribution network node 21; Microgrid 4, with 15 nodes and containing 105MW diesel generators, 35MW energy storage, and 100MW of renewable energy, is connected to distribution network node 23; and Microgrid 5, with 7 nodes and containing 85MW diesel generators, 70MW energy storage, and 300MW of renewable energy, is connected to distribution network node 31. These microgrids collectively constitute a highly pervasive renewable energy-dominated system, including microgrids exhibiting three operating states: power surplus export-oriented microgrids, power shortage load demand-oriented microgrids, and self-sufficient balancing microgrids.

[0161] Then, taking microgrid 1 as an example, various scheduling models were executed, and the resulting costs are shown in Table 2. Each cost category represents the average cost obtained under each discrete P2P transaction price. Table 2 shows that, compared with models ignoring carbon-green certificate coupling market costs, considering only carbon market costs, and considering only green certificate market costs, the low-carbon economic energy management model in this invention improves environmental benefits through the combined carbon-green certificate market cost, reducing total costs by 20.85%, 4.72%, and 5.29%, respectively. This indicates that the microgrid low-carbon economic energy management model proposed in this invention has greater economic advantages.

[0162] Table 2. Scheduling results for each model

[0163]

[0164] Finally, for this 5-microgrid interconnection system, the P2P transaction results of the system were solved using a centralized method, the traditional ADMM algorithm, and the two-round communication adaptive fast clearing method proposed in this invention. The calculation results under two market rules—without considering network usage fees and with considering network usage fees—are shown in Tables 3 and 4, respectively. Table 3 shows that, without considering network usage fees, the proposed method, through a two-round communication non-iterative calculation clearing method, reduces the computation time to 1.78% of ADMM and 32.14% of the centralized method, with a cost error of only 0.04%. Furthermore, by implementing the method in a distributed manner based on marginal utility functions, the privacy of the microgrid participants is protected. Table 4 shows that, considering network usage fees, the proposed method, through a two-round communication fast iterative clearing method, requires only 3 iterations and completes the process within 1.46 seconds. Compared with the popular distributed ADMM, the number of iterations and computation time are reduced by 98.08% and 97.72%, respectively. These calculation results all demonstrate that the method proposed in this invention has a greater computational advantage when clearing micro-network peer-to-peer transactions.

[0165] Table 3 shows the calculation results without considering network usage fees.

[0166]

[0167] Table 4 shows the calculation results considering network usage fees.

[0168]

[0169] Compared with existing technologies, this invention can balance the economic efficiency of trading schemes with the security of resisting uncertainty, and improve the computational efficiency and subject privacy of distributed trading. First, a joint opportunity-constrained microgrid energy management model is established, taking into account the joint market costs of carbon-green certificates and the adjustment of reserve demand. This model reflects the coupling relationship between carbon-green certificates and electricity through carbon quota constraints and green certificate trading constraints, and models the uncertainty-related reserve constraints and network security constraints as joint opportunity constraints that balance economic efficiency and security. Then, to efficiently achieve distributed clearing in the inter-microgrid P2P trading market, an adaptive fast distributed peer-to-peer clearing method is proposed. This method considers the complex decision-making behavior of microgrid subjects, achieving peer-to-peer clearing in a non-iterative manner when network usage fees are not considered, or in a shock-resistant fast iterative method when network usage fees are considered. This allows for peer-to-peer trading between multiple microgrid systems based on market demand and the microgrid's optimal trading intentions, improving computational efficiency and the privacy of information within the microgrid in a distributed manner.

[0170] Compared to traditional microgrid peer-to-peer trading methods, this invention has significant advantages in the following aspects: In terms of multi-microgrid peer-to-peer trading models, the coupling relationship between carbon quota constraints and green certificate trading constraints is reflected through carbon quota constraints and green certificate trading constraints, taking into account the joint benefits of the electricity-carbon-green certificate coupled market, thereby improving the operational economy of multi-microgrid systems; at the same time, the impact of the safety reserve requirements of microgrids in response to the uncertainty of renewable energy output on the willingness to engage in peer-to-peer energy trading is considered. By modeling reserve constraints and network security constraints based on affine strategies as joint opportunity constraints, the security of microgrid operation itself during peer-to-peer trading is improved. Regarding the calculation method for microgrid peer-to-peer transaction clearing, a piecewise marginal utility function of the microgrid is constructed based on an optimization method, reflecting the optimal trading intentions of microgrid entities under different peer-to-peer market prices. Then, based on this piecewise marginal utility function, an adaptive fast distributed peer-to-peer clearing method applicable to different peer-to-peer transaction market rules is proposed. This method can achieve peer-to-peer transaction clearing between microgrids in a non-iterative manner when network usage fees are ignored. When network usage fees are considered, a fast iterative method with anti-oscillation is used to achieve peer-to-peer transaction clearing between microgrids, significantly improving the distributed computing speed of peer-to-peer transactions, meeting the needs of real-time transactions and operation, and effectively protecting the information privacy of each microgrid entity.

[0171] Another embodiment of the present invention provides a local computing device for generating a microgrid marginal utility function, characterized in that the device is deployed on the microgrid side and includes at least one processor and a memory; the memory stores a computer program, which, when executed by the processor, performs the following steps:

[0172] Obtain internal resource parameters, operating cost parameters, and external market price parameters of the microgrid;

[0173] Based on the parameters, an optimal trading volume model is constructed that takes into account internal resource constraints, operating costs, and coupling with external related markets.

[0174] The optimal trading volume model is solved in parallel under multiple discrete peer-to-peer trading prices, resulting in a series of optimal trading volumes;

[0175] Based on the peer-to-peer transaction price and the corresponding optimal transaction volume, a piecewise linear marginal utility function for the microgrid is generated; the piecewise linear marginal utility function is output and uploaded to the global market controller.

[0176] In some embodiments, external market price parameters include carbon emission rights trading prices and green certificate trading prices.

[0177] In some embodiments, the optimal trading volume model is an opportunity-constrained optimization model that takes into account the uncertainty of new energy output.

[0178] Another embodiment of the present invention provides a distributed resource peer-to-peer trading system, characterized in that it includes:

[0179] The local computing device of the present invention corresponds to a microgrid for each local computing device.

[0180] The Global Market Controller (GMC), which communicates with all local computing devices, is configured to: receive piecewise linear marginal utility functions uploaded by all microgrids; construct a market clearing model based on all piecewise linear marginal utility functions, transaction balance constraints, and network security constraints; solve the market clearing model to obtain peer-to-peer transaction clearing results; and broadcast the clearing results to the local computing devices corresponding to each microgrid.

[0181] In some embodiments, GMC is further configured to: solve the market clearing model using the iterative algorithm of the present invention when considering network usage fees.

[0182] Another embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the steps of the method of the present invention.

[0183] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0184] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0185] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0186] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0187] The background section of this invention may include background information about the problems or environment in which the invention is being developed, and is not necessarily a description of prior art. Therefore, the content included in the background section does not constitute an admission of prior art by the applicant.

[0188] The above description provides a further detailed explanation of the present invention in conjunction with specific / preferred embodiments, and it should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various substitutions or modifications can be made to these described embodiments without departing from the concept of the present invention, and all such substitutions or modifications should be considered within the scope of protection of the present invention. In the description of this specification, the reference to terms such as "an embodiment," "some embodiments," "preferred embodiment," "example," "specific example," or "some examples," etc., indicates that the specific features, structures, materials, or characteristics described in connection with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any suitable manner in one or more embodiments or examples. Without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification and the features of different embodiments or examples. Although the embodiments of the present invention and their advantages have been described in detail, it should be understood that various changes, substitutions, and modifications can be made herein without departing from the scope of protection of the patent application.

Claims

1. A microgrid adaptive distributed point-to-point collaborative method integrating multiple market constraints, characterized in that, Includes the following steps: In each microgrid, based on its internal resource constraints, operating costs, and coupling relationship with external related markets, an optimal transaction volume model is constructed for each microgrid under different peer-to-peer transaction prices; The optimal trading volume under multiple point-to-point trading prices is calculated in parallel, and functional relationship data reflecting the trading intentions of each microgrid under the trading price is generated. Each microgrid uploads the aforementioned functional relationship data to the Global Market Controller (GMC); The GMC receives the functional relationship data of all microgrids and constructs a market clearing model based on it. The market clearing model aims to maximize system benefits or minimize total costs and satisfies transaction balance constraints and network security constraints. The GMC solves the market clearing model to obtain the final peer-to-peer transaction clearing result, and broadcasts the clearing result to each microgrid. The objective function of the optimal trading volume model includes the costs or benefits arising from participating in the carbon emission trading market and the green certificate trading market, and the optimal trading volume model takes into account the uncertainty of new energy output, and models the backup constraints and / or cybersecurity constraints related to uncertainty as opportunity constraints. The functional relationship data is a piecewise linear marginal utility function with transaction price as the independent variable and optimal transaction volume as the dependent variable, obtained by performing piecewise linear fitting on the multiple point-to-point transaction prices and the calculated optimal transaction volume. The market clearing model is a mixed-integer linear programming model. It introduces auxiliary integer variables to handle the nonlinear relationships caused by identifying effective constraints after incorporating the piecewise linear marginal utility function into the market clearing model.

2. The microgrid adaptive distributed point-to-point collaborative method integrating multiple market constraints as described in claim 1, characterized in that, The internal resource constraints include at least one of the following: output constraints, ramping constraints, and standby constraints of diesel generators; output constraints of new energy units; and output constraints and standby constraints of energy storage devices. The coupling relationship with external related markets includes the coupling relationship with the power grid's electricity trading market, carbon emission trading market, green certificate trading market, and backup ancillary service market.

3. The microgrid adaptive distributed point-to-point collaborative method integrating multiple market constraints as described in claim 1, characterized in that, Taking network usage fees into account, the GMC employs an iterative algorithm to solve the market clearing model, the iterative algorithm including: Based on the current equivalent transaction price, solve the first optimization problem with the goal of maximizing system benefits and satisfying transaction balance and network security constraints to obtain the optimal transaction volume for each microgrid; fix the transaction volume and solve the second optimization problem with the goal of minimizing network usage fees to obtain the specific transaction volume distribution among microgrids and calculate the network usage fees; update the equivalent transaction price for each microgrid according to the network usage fees; iterate until convergence. Without considering network usage fees, the GMC employs a non-iterative algorithm to solve the market clearing model, including the following steps: constructing a market clearing model with the goal of maximizing system efficiency, the decision variables of which include the peer-to-peer transaction volume of each microgrid and the transaction volume with the upper-level grid; converting the functional relationship data uploaded by each microgrid into a mixed-integer linear programming model to characterize the piecewise linear relationship between transaction volume and price; solving the mixed-integer linear programming model to obtain the final peer-to-peer transaction clearing price and the clearing transaction volume of each microgrid in one step.

4. The microgrid adaptive distributed point-to-point collaborative method integrating multiple market constraints as described in claim 3, characterized in that, The iterative algorithm also includes an oscillation monitoring and correction step: when an oscillation is detected in the update of the equivalent transaction price, a binary search method is used to search for the optimal equivalent transaction price increment within the oscillation range.

5. A local computing device for generating the marginal utility function of a microgrid, characterized in that, The device is deployed on the microgrid side and includes at least one processor and a memory; the memory stores a computer program, which, when executed by the processor, performs the following steps: Obtain internal resource parameters, operating cost parameters, and external market price parameters of the microgrid; Based on the above parameters, an optimal trading volume model is constructed that considers internal resource constraints, operating costs, and coupling with external related markets. The objective function of the optimal trading volume model includes the costs or benefits incurred from participating in the carbon emission trading market and the green certificate trading market. Furthermore, the optimal trading volume model considers the uncertainty of new energy output and models the backup constraints and / or cybersecurity constraints related to uncertainty as opportunity constraints. The optimal trading volume model is solved in parallel under multiple discrete peer-to-peer trading prices to obtain a series of optimal trading volumes; Based on the point-to-point transaction price and the corresponding optimal transaction volume, functional relationship data of the microgrid is generated. The functional relationship data reflects the transaction intention of each microgrid at the transaction price. The functional relationship data is a piecewise linear marginal utility function with transaction price as independent variable and optimal transaction volume as dependent variable, obtained by piecewise linear fitting of the multiple point-to-point transaction prices and the calculated optimal transaction volume. The functional relationship data is output and uploaded to the Global Market Controller (GMC). The functional relationship data is used to construct a market clearing model. The market clearing model aims to maximize system efficiency or minimize total cost and satisfies transaction balance constraints and network security constraints. The market clearing model is a mixed-integer linear programming model. It introduces auxiliary integer variables to handle the nonlinear relationships caused by identifying effective constraints after incorporating the piecewise linear marginal utility function into the market clearing model.

6. The local computing device according to claim 5, characterized in that, The external market price parameters include carbon emission rights trading prices and green certificate trading prices, and the optimal trading volume model is an opportunity-constrained optimization model that takes into account the uncertainty of new energy output.

7. A distributed resource peer-to-peer trading system, characterized in that, include: Multiple local computing devices as described in claim 5 or 6, each of the local computing devices corresponding to a microgrid; The Global Market Controller (GMC), communicatively connected to all the local computing devices, is configured to: receive function relationship data uploaded by all microgrids; construct a market clearing model based on all function relationship data, transaction balance constraints, and network security constraints; solve the market clearing model to obtain peer-to-peer transaction clearing results; wherein, when considering network usage fees, the GMC uses the microgrid adaptive distributed peer-to-peer collaborative method as described in claim 3 or 4 to solve the market clearing model; and broadcast the clearing results to the local computing devices corresponding to each microgrid.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 4.

Images