A robust optimization scheduling method and system for multi-micronet participation in electricity auxiliary market

Abstract

The application relates to the technical field of power dispatch, in particular to a robust optimization dispatching method and system for multiple microgrids participating in an electricity auxiliary market. The method comprises the following steps: constructing a three-layer dispatching architecture; establishing a multi-market clearing model based on a supply-side price quota curve and a demand-side price quota curve; selecting the supply-side price quota curve or the demand-side price quota curve according to the net load state of the microgrid in the next time period t to participate in the bidding of the electricity auxiliary market, and obtaining the market result; establishing a power distribution optimization dispatching model of the active distribution network layer, wherein the power distribution optimization dispatching model takes minimizing the total operation cost of the distribution network system as the target; respectively establishing an optimization dispatching model of each microgrid, wherein the optimization dispatching model of each microgrid takes minimizing the operation cost of the microgrid as the target; and generating a real-time dispatching strategy of the microgrid according to the optimization dispatching model of each microgrid, and executing the real-time dispatching strategy of the microgrid in the next time period t.

Description

Technical Field

[0001] This application relates to the field of power dispatching technology, specifically to a robust optimization dispatching method and system for multiple microgrids participating in the auxiliary power market. Background Technology

[0002] With the rapid development of renewable energy and the large-scale integration of distributed energy resources, microgrids, as regional integrated energy systems that integrate distributed photovoltaic, wind power, energy storage systems, and controllable loads, are playing an increasingly important role in improving energy utilization efficiency, enhancing power supply reliability, and promoting the consumption of clean energy. Multiple microgrids interconnect to form multi-microgrid systems can achieve coordinated, complementary, and optimized energy allocation over a wider area. However, due to the significant intermittency and uncertainty of renewable energy output, coupled with frequent fluctuations in user-side loads, the operation and scheduling of multi-microgrid systems face enormous challenges. To enhance the enthusiasm and economic efficiency of multi-microgrids participating in the electricity market, the ancillary services market (AS / RS) has been gradually opened up in recent years, encouraging distributed resources to participate in market bidding and clearing in an aggregated manner. The active distribution network, as the intermediate hub connecting the upper-level grid and the lower-level microgrids, plays a crucial role in acting as an intermediary for multiple microgrids participating in AS / RS transactions.

[0003] Traditional centralized dispatching models are ill-suited to market environments characterized by multiple stakeholders, multiple objectives, and high uncertainty. Furthermore, they suffer from shortcomings in areas such as market clearing mechanism modeling, multi-level interest coordination, and real-time response capabilities. Especially in the face of fluctuating electricity prices, load forecasting errors, and increasingly stringent carbon emission constraints, constructing a multi-microgrid collaborative dispatching mechanism that balances economic efficiency, robustness, and low carbon footprint has become a key technical challenge for the development of smart distribution networks. Summary of the Invention

[0004] This specification describes a robust optimization scheduling method and system for multi-microgrid participation in the electric auxiliary market through several embodiments.

[0005] Firstly, embodiments of this specification provide a robust optimization scheduling method for multiple microgrids participating in the electric vehicle auxiliary power market, including the following steps:

[0006] A three-layer dispatch architecture is constructed, consisting of an upper-layer auxiliary power market clearing layer, a middle-layer active distribution network layer, and a lower-layer multi-microgrid layer. The active distribution network layer acts as an agent for multiple microgrids in the lower layer to bid and clear in the auxiliary power market with the upper-level power grid, and converts the market results into incentive prices and distributes them to each microgrid.

[0007] A multi-market clearing model is established based on the supply-side price quota curve and the demand-side price quota curve;

[0008] The active distribution network layer selects either the supply-side price quota curve or the demand-side price quota curve to participate in the bidding for the auxiliary power market based on the net load status of the microgrid in the next time period t, and obtains the market results, which include the cleared power volume and the cleared power price.

[0009] A distribution optimization scheduling model for the active distribution network layer is established, with the objective of minimizing the total operating cost of the distribution network system.

[0010] Optimization scheduling models for each microgrid are established, with the goal of minimizing the operating cost of the microgrid.

[0011] Each microgrid generates its own real-time scheduling strategy based on its optimized scheduling model, which is then executed by the microgrid in the next time period t.

[0012] Secondly, embodiments of this specification provide a robust optimized scheduling system for multi-microgrid participation in the electric vehicle auxiliary power market, comprising:

[0013] The architecture module constructs a three-layer scheduling architecture, including an upper-layer auxiliary power market clearing layer, a middle-layer active distribution network layer, and a lower-layer multi-microgrid layer. The active distribution network layer acts as an agent for multiple microgrids in the lower layer to bid and clear in the auxiliary power market with the upper-level power grid, and converts the market results into incentive prices and distributes them to each microgrid.

[0014] The model module establishes a multi-market clearing model based on the supply-side price quota curve and the demand-side price quota curve;

[0015] The bidding module allows the active distribution network layer to select either the supply-side price quota curve or the demand-side price quota curve to participate in the bidding for the auxiliary power market based on the net load status of the microgrid in the next time period t, and obtain the market results, which include the cleared power volume and the cleared power price.

[0016] The first optimization module establishes a distribution optimization scheduling model for the active distribution network layer, with the goal of minimizing the total operating cost of the distribution network system.

[0017] The second optimization module establishes an optimization scheduling model for each microgrid, with the goal of minimizing the operating cost of the microgrid.

[0018] The scheduling module generates real-time scheduling strategies for each microgrid based on its own optimized scheduling model, which are then executed by the microgrid in the next time period t.

[0019] Thirdly, embodiments of this specification provide an electronic device, including a processor and a memory;

[0020] The processor is connected to the memory;

[0021] The memory is used to store executable program code;

[0022] The processor runs a program corresponding to the executable program code stored in the memory to perform the method described in any of the above aspects.

[0023] Fourthly, embodiments of this specification provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the methods described in any of the above aspects.

[0024] Fifthly, embodiments of this specification provide a computer program product, including a computer program that, when executed by a processor, implements the methods described in any of the above aspects.

[0025] The beneficial effects of the technical solutions provided in some embodiments of this specification include at least the following:

[0026] In several embodiments of this specification, the robust optimization scheduling method and system for multi-microgrid participation in the auxiliary power market provides a three-layer scheduling architecture. The active distribution network acts as an agent for lower-level microgrids to participate in the bidding and clearing of the auxiliary power market, and the market results are converted into incentive price distribution, improving market clearing speed and microgrid response efficiency. A multi-market clearing model is established based on supply-side and demand-side price quota curves. By discretizing bid segments, introducing state label variables and continuous transaction volume variables, the clearing price and clearing volume are calculated, enhancing the adaptability and executability of the scheduling strategy in the actual market environment. A two-stage robust optimization model is used for microgrid scheduling. The first stage generates a day-ahead scheduling plan based on day-ahead forecasts, and the second stage generates a real-time scheduling strategy by combining real-time forecast information, effectively addressing the uncertainty of renewable energy output and load demand. It supports microgrids in conducting point-to-point power trading while receiving incentive prices, optimizing local energy resource allocation, improving the local consumption level of renewable energy, and enhancing the autonomy and flexibility of the multi-microgrid system.

[0027] Other features and advantages of various embodiments of this specification will be further revealed in the following detailed description and accompanying drawings. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of this specification, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1This is a schematic diagram of the multi-microgrid system architecture used in this specification.

[0030] Figure 2 This is a schematic diagram of the robust optimization scheduling method provided in this specification.

[0031] Figure 3 This is a schematic diagram of the power node architecture used in this specification.

[0032] Figure 4 The SPQC and DPQC curve diagrams provided in this manual are shown.

[0033] Figure 5 This is a diagram illustrating the market results provided in this specification.

[0034] Figure 6 This document provides a diagram illustrating the cleared electricity volume and price in the energy market.

[0035] Figure 7 This document provides a diagram illustrating the clearing of reactive power and pricing in the reactive power auxiliary power market.

[0036] Figure 8 This diagram illustrates the cleared electricity volume and price in the peak-shaving auxiliary market provided in this manual.

[0037] Figure 9 This is a schematic diagram of the carbon quota and carbon emission curve provided in this specification.

[0038] Figure 10 This diagram illustrates the operating costs and safety slack margins of the MMGs provided in this manual.

[0039] Figure 11 This is a schematic diagram of MG1 output with and without considering point-to-point energy transactions, as provided in this manual.

[0040] Figure 12 This is a schematic diagram of MG2 output with and without considering point-to-point energy trading, as provided in this manual.

[0041] Figure 13 This is a schematic diagram of MG3 output with and without considering point-to-point energy trading, as provided in this manual.

[0042] Figure 14 This is a schematic diagram of the electronic device provided in this manual. Detailed Implementation

[0043] The technical solutions of the embodiments of this specification will be explained and described below with reference to the accompanying drawings. However, the following embodiments are only preferred embodiments of this specification and not all of them. Other embodiments obtained by those skilled in the art based on the embodiments in the implementation methods without creative effort are all within the protection scope of this specification.

[0044] The terms "first," "second," "third," etc., in the description, claims, and accompanying drawings are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or apparatus.

[0045] In the following description, terms such as “inner,” “outer,” “upper,” “lower,” “left,” and “right” are used only to facilitate the description of the embodiments and to simplify the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this specification.

[0046] All data involved in this application are information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0047] Before introducing the technical solutions described in this manual, the application scenarios and related technologies of the technical solutions will be introduced.

[0048] The increasing proportion of renewable energy sources, such as wind and solar power, in the power system poses a significant challenge to the safe and stable operation of the power grid due to the randomness and volatility of their output. Microgrids (MGs), as small-scale autonomous systems integrating distributed renewable energy, energy storage, and loads, have become a key technology for achieving local consumption of renewable energy. Multiple microgrids are further interconnected to form multi-microgrid systems (MMGs), and researching their trading strategies for participating in the electricity market is of great significance for improving the system's economic efficiency and operational reliability.

[0049] Multi-microgrid (MMG) systems are typically modeled as directly trading energy with the upper-level grid. However, in actual operation, the power balancing capability of microgrids is constrained by the limited installed capacity of conventional distributed generation and the intermittent output of renewable energy. To ensure reliable power supply to critical loads, microgrids need to be interconnected with the distribution network (DN) through a point of common coupling (PCC). This makes the distribution system operator (DSO) a crucial hub for interaction between MMGs and the upper-level grid. Therefore, establishing a collaborative optimization scheduling mechanism between the DSO and MMGs is essential.

[0050] The strong uncertainty inherent in renewable energy significantly increases the risks for multi-microgrid (MMG) systems participating in market transactions and dispatch optimization. Currently, the mainstream methods for handling such uncertainty are stochastic optimization and robust optimization. While stochastic optimization is more economical, its solution relies on the precise probability distribution of uncertain parameters, which is difficult to obtain in practice. In contrast, robust optimization only requires setting the uncertainty set, and its solution is feasible under all uncertainty scenarios, making it more applicable. Although existing research applies robust optimization to handle uncertainty, it generally treats MMG systems as passive recipients of market outcomes and dispatch optimization results, failing to fully explore their potential as active participants in dispatch optimization with DSOs and other MGs.

[0051] Therefore, this specification provides a market architecture with distribution network operators as the hub, including a three-layer dispatch architecture: an upper-layer auxiliary power market clearing layer 101, a middle-layer active distribution network layer 103, and a lower-layer multi-microgrid layer 104, as shown in the appendix. Figure 1 As shown, this enables distribution network operators to represent multi-microgrid (MMG) systems in bidding for electricity and ancillary services in the upstream market, and to guide the optimized operation of MMG systems by issuing incentive tariffs downstream, thereby fully leveraging the coordinating role of distribution network operators.

[0052] First, this specification provides a robust optimization scheduling method for multi-microgrid participation in the electric vehicle auxiliary power market. Please refer to the appendix. Figure 2 The steps include:

[0053] Step S1) Construct a three-layer dispatch architecture including an upper-layer auxiliary power market clearing layer 101, a middle-layer active distribution network layer 103, and a lower-layer multi-microgrid layer 104. The active distribution network layer 103 acts as an agent for multiple microgrids in the lower layer to bid and clear in the auxiliary power market with the upper-level power grid, and converts the market results into incentive prices and distributes them to each microgrid.

[0054] The upper-level auxiliary power market clearing layer 101 is responsible for receiving bidding information from various market participants and completing clearing calculations according to market rules, determining a unified clearing price and clearing volume. The middle-level active distribution network layer 103 acts as an aggregator and dispatcher for multiple microgrids in the lower layer, collecting information on the generation capacity, load demand, and regulation potential of each microgrid. (See attached...) Figure 3 As shown, the middle-layer active distribution network 103 represents all subordinate microgrids in submitting unified bids and participating in the clearing process in the upper-level auxiliary power market. After obtaining the market clearing results, the active distribution network 103 transforms information such as the clearing price and allocated transaction volume into incentive price signals for each microgrid. This incentive price comprehensively reflects market revenue and system operation requirements, guiding each microgrid to rationally allocate its internal resources for output and consumption. Based on the received incentive price and its own operational constraints, each lower-layer microgrid autonomously optimizes the scheduling of its internal distributed power sources, energy storage devices, and controllable loads. This achieves distributed coordination in scheduling, improves the efficiency of multi-microgrid systems participating in the auxiliary power market, and provides conditions for subsequent scheduling optimization.

[0055] Step S2) Establish a multi-market clearing model based on the supply-side price quota curve and the demand-side price quota curve.

[0056] Specifically, it includes:

[0057] The supply-side price quota curve and the demand-side price quota curve are each converted into several discrete price segments, as shown in the appendix. Figure 4 As shown, the bidding segment records the bid price and the corresponding maximum tradable electricity volume;

[0058] A status flag is introduced to indicate whether the supply-side price quota curve and the price quota curve segment of the demand-side price quota curve are fully referenced in the market clearing model solution;

[0059] For a quoted price segment that is referenced but not fully referenced, a variable with continuous values ​​is set to represent the electricity volume of the quoted price segment referenced by the market clearing model, which is denoted as the transaction electricity volume.

[0060] The market-clearing price is determined by the intersection of total market supply and total demand, and the estimated bidding strategy is obtained based on the historical bidding strategies of the active distribution network layer 103.

[0061] A market clearing model is constructed. The input of the market clearing model is the estimated bidding strategy of the active distribution network layer 103, and the output is the clearing price and the clearing volume.

[0062] Discretization modeling transforms continuous bidding behavior into a computable piecewise step function, facilitating mathematical expression and solution within the optimization model. A binary state flag variable is introduced to characterize whether each bid segment is fully activated or partially referenced during market clearing. For each bid segment, this state flag indicates its market participation status: a flag of 0 indicates the bid segment is not selected; a flag of 1 indicates the bid segment is fully or partially used. For bid segments that are referenced but not fully traded (i.e., bid segments at the marginal transaction price), a continuous variable is set to represent the actual electricity volume adopted by the market, denoted as the traded electricity volume. The market clearing price is determined according to the supply-demand equilibrium principle, i.e., the clearing price is determined by the intersection of the cumulative supply curve on the supply side and the cumulative demand curve on the demand side. Only when the trading of a bid segment causes the aggregate supply and aggregate demand to reach equilibrium for the first time, the corresponding electricity price is the final market clearing price. Therefore, the clearing price depends on the price level of the last bid segment that is partially or fully activated.

[0063] Based on historical bidding data and operational experience of the active distribution network layer 103, its bidding behavior patterns are analyzed to construct a predictive bidding strategy. This allows for the acquisition of meaningful market predictions even without receiving bidding strategies from other auxiliary market participants. Specifically, the market-clearing price is determined by the intersection of total supply and total demand, serving as the prediction of the final market outcome. The accuracy of this prediction meets the requirements of the technical solution described in this specification. The prediction is based on historical bidding data because the load and generation output of the microgrid exhibit short-term continuity and stability.

[0064] The market-cleared electricity volume and price are determined by the following formula:

[0065]

[0066]

[0067] in and These are the clearing price and the bidding price, respectively. and These are the cleared electricity volume and the bidding electricity volume, respectively. F x and F λ The clearing process for market electricity volume and market electricity price is determined by market clearing rules and the bidding parameters of competitors.

[0068] Step S3) The active distribution network layer 103 selects either the supply-side price quota curve or the demand-side price quota curve to participate in the bidding for the ancillary services market based on the net load status of the microgrid in the next time period t, and obtains the market results, which include the cleared electricity volume and the cleared electricity price. By introducing Boolean variables and piecewise linearization methods, the clearing results of the electricity market and the ancillary services market are obtained.

[0069] The specific methods for obtaining market results include: reading the predicted load and predicted power generation of each microgrid in the next time period t, calculating the net load of each microgrid, determining the market role of each microgrid based on the net load of each microgrid, and then selecting the supply-side price quota curve or the demand-side price quota curve and inputting it into the market clearing model.

[0070] Based on the transformed supply-side price quota curve or demand-side price quota curve, and the estimated bidding strategy, the market clearing model is solved to obtain the intersection point, and the market outcome is obtained based on the intersection point.

[0071] Subtracting the predicted generation output from the predicted load yields the net load for each microgrid. A positive net load indicates that the microgrid may need to purchase electricity from external sources to meet its demand; conversely, a negative net load suggests that the microgrid may sell electricity to the external market. Demand-side participants use demand-side price quota curves to participate in the market. Supply-side participants use supply-side price quota curves for bidding. Boolean variables are introduced to indicate whether each bid segment is selected for market clearing. Furthermore, for partially referenced bid segments, continuous decision variables are used to represent the traded electricity volume. Piecewise linearization is used to approximate the supply-side and demand-side price quota curves, transforming the complex nonlinear optimization problem into an easily solvable mixed-integer linear programming (MILP) problem. The selected supply-side or demand-side price quota curve and the estimated bidding strategy are input into the market clearing model. Based on these inputs, the market clearing model is solved to find the intersection point where aggregate supply equals aggregate demand; this intersection point determines the market clearing price.

[0072] Step S4) Establish a distribution optimization scheduling model for the active distribution network layer 103, wherein the distribution optimization scheduling model aims to minimize the total operating cost of the distribution network system.

[0073] While receiving the incentive electricity price issued by the active distribution network layer 103, the microgrid conducts point-to-point electricity transactions with other microgrids.

[0074] The total operating cost of the distribution network system includes the operation and maintenance costs of the micro gas turbines, the operation and maintenance costs of renewable energy, the charging, discharging and driving costs of the electric vehicle cluster, the revenue from energy trading with the microgrid, market clearing fees, and carbon trading costs. The total operating cost of the distribution network system can be calculated using the methods described later in this specification, or other methods known in the art as needed. The content described in this specification does not constitute a limitation on the method of calculating the total operating cost of the distribution network system.

[0075]

[0076] Among them, C MT The operating and maintenance cost of a micro gas turbine (MT) (a second-order function); C WT and C PV For the operation and maintenance costs of wind turbines and solar power; C EV The charging, discharging, and driving costs of electric vehicle fleets; C EX For the revenue from energy trading between the active distribution network layer 103 and MMGs; C clear C. Market clearing costs; C This refers to the cost of carbon trading.

[0077] The constraints of the power distribution optimization scheduling model include power flow constraints, upper and lower limits for distributed generation output, electric vehicle charging and discharging logic and energy state constraints, upper and lower limits for incentive electricity prices, and carbon emission quota and actual emission calculation constraints. The constraints of the active distribution network layer 103 include:

[0078] Distribution network power flow constraints ensure power balance and node voltage safety in the radial network. Its mathematical expression is:

[0079]

[0080] Where ψ is the set of nodes in the system, ε is the set of branches in the system; P ij and Q ij P represents the active and reactive power flowing from node i to node j; i and Q i Represents the active and reactive power of node i and node j; and This represents the active and reactive power generation of node j; and This represents the active and reactive power consumption of node j. To ensure power balance at nodes, the distribution network must meet power constraints at node i. Node 1 is defined as the node representing the Distribution System Operator (DSO) that engages in energy trading with the upstream grid. In this specification, the DSO plays the same role as the active distribution network layer 103.

[0081]

[0082]

[0083] in, and This represents the active and reactive load demand of the i-th node. Simultaneously, the voltage at node i should satisfy the following constraints:

[0084]

[0085] Output limits for micro gas turbines (MT) and renewable energy sources (RES):

[0086]

[0087]

[0088] in, P is the power factor of the gas turbine. WT,pre and P PV,pre This is a predicted value for the scenery.

[0089] The charging and discharging logic, state of energy, and driving constraints of the electric vehicle (EV) cluster include:

[0090]

[0091]

[0092]

[0093] in, Let be the energy (MWh) of the electric vehicle at node i at time t. and This indicates the charging / discharging status of the electric vehicle. A value of 0 indicates that no charging / discharging is currently in progress, while a value of 1 indicates that charging / discharging is currently in progress. and This indicates the charging / discharging status of an electric vehicle while it is being driven, and it is stipulated that charging / discharging is prohibited while driving. and Let i be the initial / final energy storage of the electric vehicle at node i.

[0094] The mathematical expression for the upper and lower limits of the incentive electricity price constraints issued to the lower-level MG is:

[0095]

[0096] Carbon trading constraints, including the calculation of carbon emission allowances and actual emissions, are expressed mathematically as follows:

[0097]

[0098] Step S5) Establish an optimal scheduling model for each microgrid, with the goal of minimizing the operating cost of the microgrid.

[0099] While receiving incentive tariffs from the active distribution network layer 103, the microgrid engages in point-to-point electricity trading with other microgrids. The operating cost of the microgrid's optimized scheduling model includes the operation and maintenance costs of the energy storage system, the energy interaction costs with the active distribution network layer 103, and the point-to-point transaction costs. The operating cost of the microgrid can be calculated using the schemes described later in this specification, or other methods known in the art can be used as needed. The content described in this specification does not constitute a limitation on the method of calculating the operating cost of the microgrid.

[0100] A scheduling optimization model for lower-level multi-microgrids (MMGs) is established, with each MG aiming to minimize its own operating cost. Each MG optimizes its internal resource scheduling based on the incentive price issued by the active distribution network layer 103 and the point-to-point transaction situation. Its objective function is:

[0101]

[0102] Among them, C ESS For the operation and maintenance costs of energy storage, C DT Costs of direct transactions between MGs.

[0103] The constraints of the microgrid's optimal scheduling model include active power balance constraints, reactive power balance constraints, distributed generation output constraints, energy storage system charging and discharging power and energy state constraints, point-to-point bilateral trading balance constraints, and carbon trading constraints. Specific constraints are as follows.

[0104] The mathematical expression for the power-reactive power balance constraint is:

[0105]

[0106] The output constraints of micro gas turbines (MT) and renewable energy sources (RES) are mathematically expressed as follows:

[0107]

[0108]

[0109] The mathematical expressions for the charging and discharging power and energy state constraints of an energy storage system (ESS) are as follows:

[0110]

[0111]

[0112] The bilateral balance constraint in peer-to-peer transactions ensures transaction equivalence, and its mathematical expression is:

[0113]

[0114] Among them, when >0 indicates that the m-th MG buys electricity from the n-th MG. <0 indicates that the m-th MG sells electricity to the n-th MG. =0 indicates that there is no direct peer-to-peer transaction between the two MGs. On the other hand, carbon trading constraints can also be included, the mathematical expression of which is:

[0115]

[0116]

[0117] Step S6) Each microgrid generates a real-time scheduling strategy based on its own optimal scheduling model, which is then executed by the microgrid in the next time period t. Robust optimization Mox is used to handle the uncertainty of wind and solar power output, and KKT conditions and strong duality theory are used for efficient solution.

[0118] To address the stochastic nature of wind and solar power output, this manual establishes a robust min-max optimization model with an uncertainty set. The first stage involves decision-making regarding market bidding and day-ahead scheduling; the second stage adjusts the real-time scheduling strategy under the worst-case wind and solar power scenario. To solve this complex multi-layered optimization model, this manual provides the following steps:

[0119] The optimization problems of each lower-level MG are replaced with their KKT optimality conditions (including primal feasibility, dual feasibility, and complementary relaxation conditions); a strong duality theorem is introduced to transform the nonlinear complementary relaxation conditions into equivalent linear constraints; the original two-level robust optimization model is transformed into a single-level mathematical programming (MPEC) problem with equilibrium constraints; this mathematical programming problem is transformed into a standard mixed integer linear programming (MILP) model and solved using a publicly available solver.

[0120] To effectively address the inherent volatility and uncertainty of renewable energy output such as photovoltaic and wind power, this method employs a robust optimization framework for modeling. Specifically, a two-stage robust optimization model with an uncertainty set, consisting of a min-max stage, is constructed. The first stage is the "day-ahead decision stage," where the microgrid makes preliminary decisions based on forecast data, including market bidding, generator start-up and shutdown, and energy storage charging and discharging plans, thus forming a day-ahead dispatch plan. The second stage is the "real-time adjustment stage," where, under the worst-case scenario known in history but still within the uncertainty set of wind and solar power output, controllable variables such as power generation and energy storage capacity are readjusted to enhance the adaptability of the dispatch scheme and the system's robustness under extreme conditions.

[0121] Because the model has a two-layer structure of "upper-level decision-making - lower-level response" and the objective function contains minima-maxima nesting, direct solution is difficult and computationally complex. Therefore, this specification proposes an efficient mathematical transformation method to achieve model solvability. First, the real-time scheduling optimization problem of each microgrid in the lower layer (i.e., the second-stage problem) is replaced with its Karush-Kuhn-Tucker (KKT) optimality conditions. KKT conditions include the feasibility of the original constraints, the non-negativity of the dual variables (dual feasibility), and the linear relationship between the gradient of the objective function and the gradient of the constraints. The most crucial condition is the complementary relaxation condition, which reflects the relationship between whether the constraints are effective and the corresponding dual variables.

[0122] Secondly, for the nonlinear complementary relaxation terms in the KKT conditions (i.e., the product of the dual variable and the constraint relaxation is zero), a strong duality theorem is introduced. This theorem states that, under certain regularity conditions, the optimal value of the primal problem equals the optimal value of the dual problem. Using this property, the non-convex, nonlinear complementary relaxation conditions can be equivalently transformed into a set of linear equality constraints, thereby eliminating the nonlinear terms and achieving linear reconstruction of the model. Through this transformation, the original two-level robust optimization problem is reconstructed into a single-level mathematical programming problem, containing equilibrium constraints derived from the KKT conditions, belonging to the equilibrium constraint problem (MPEC) in mathematical programming. While the MPEC model remains structurally complex, it provides a foundation for further processing.

[0123] By introducing auxiliary binary variables and the Big M method, the MPEC model is linearized in terms of the piecewise logical relationships and complementary constraints, ultimately transforming it into a standard Mixed-Integer Linear Programming (MILP) model. This MILP model possesses favorable mathematical properties and can be efficiently solved using mature open-source solvers (such as Cplex and Gurobi), yielding a real-time microgrid scheduling strategy that balances economy and robustness.

[0124] Alternatively, other publicly available solutions in this field can be used to solve the problem.

[0125] On the other hand, in another implementation, the microgrid's optimal scheduling model is a two-stage robust optimization model, which includes a first-stage robust optimization model and a second-stage robust optimization model.

[0126] The first-stage robust optimization model reads the day-ahead predicted load and day-ahead predicted generation output for each time period, calculates the day-ahead net load for each microgrid for each time period, solves the market clearing model based on the day-ahead net load, obtains the day-ahead market results, and generates the day-ahead dispatch strategy for the microgrid based on the day-ahead market results.

[0127] The second-stage robust optimization model reads the predicted load and predicted power generation for the next time period t, calculates the net load for each microgrid in the next time period t, solves the market clearing model based on the net load, obtains the market results, and generates a real-time dispatch strategy for the microgrid based on the market results.

[0128] To verify the effectiveness of the robust optimization scheduling method for MMGs participating in multi-market bidding considering DSO proposed in this specification, a detailed description is provided below with reference to the accompanying drawings and specific embodiments. This specification constructs a test case based on a modified IEEE 33-bus system, which includes an Active Distribution Network (ADN) and three connected microgrids (MGs), as shown in the figure. Figure 4 As shown.

[0129] The network structure employs a modified IEEE 33-node distribution system as the backbone of the ADN. Node 1 serves as the connection point between the ADN and the main power grid. MG 1, MG 2, and MG 3 are connected to nodes 21, 30, and 13, respectively. The system includes two electric vehicle aggregators (EVAs), each aggregating 100 electric vehicles, connected to nodes 8 and 28, respectively. The price quota curve (PQC) parameters for the energy market and ancillary services market are set according to the reference.

[0130] As a market agent, the DSO selects to participate in market bidding using either supply-side PQC (SPQC) or demand-side PQC (DPQC) based on the system's net load. Figure 5 This paper presents the bid price and market clearing price of DSOs under two different PQC (Power Quality Control) models. The results show that when a DSO acts as a seller of electricity (SPQC), the market clearing price is higher than its bid price; when it acts as a buyer of electricity (DPQC), the market clearing price is lower than its bid price. This mechanism ensures the economic viability of DSO participation in market transactions. The clearing prices under the two PQC models are essentially consistent, validating the rationality of the market clearing model. Figure 6 This demonstrates the cleared volume and price of electricity in the power market. During high-price periods (10:00-17:00), DSOs, as producers, sell electricity to the market; during low-price periods (4:00-9:00), DSOs, as consumers, purchase electricity from the market. Notably, during the high-price period from 18:00-23:00, due to the significant electricity demand of the microgrid, DSOs still choose to purchase electricity to ensure power supply reliability, reflecting their decision-making flexibility. Figure 7 The results of the reactive power ancillary service market clearing were presented. During peak load periods (5:00-9:00 and 18:00-21:00), DSOs purchased a large amount of reactive power to compensate for the system's reactive power deficit, while during off-peak load periods (10:00-17:00), DSOs purchased significantly less reactive power because the gas turbines could meet most of the reactive power demand. Figure 8 The data shows the clearing process in the peak-shaving ancillary services market. During the valley filling period (2:00-5:00), a lower clearing price incentivizes DSOs to purchase electricity; during the peak shaving period (18:00-21:00), a higher clearing price incentivizes DSOs to sell electricity; at other times, DSOs do not participate in this market transaction.

[0131] Figure 9 The carbon allowance and actual carbon emission curves for the DSO and three MGs are shown. Due to the high proportion of renewable energy in the microgrid, the carbon allowances for the DSO and each MG are higher than the actual carbon emissions, and the resulting allowance surplus can be sold for profit in the carbon trading market. As shown in Table 1, the impact of carbon price on system operation is further analyzed: when the carbon price is below $50 / ton, the gas turbines maintain full power output, and the carbon emissions remain constant; when the carbon price is above $50 / ton, the high carbon revenue prompts the gas turbines to reduce output, actively reducing carbon emissions while meeting load demand, thereby increasing carbon trading revenue.

[0132] Table 1. Costs of DSO and MG at different carbon prices

[0133]

[0134] To analyze the impact of wind and solar power output uncertainty on system scheduling, different robust adjustment parameters (γ) were set. Table 2 shows the DSO cost, MMGs cost, and total social benefit under different robust parameters. The results show that as the robust parameter increases (the degree of wind and solar power uncertainty increases), the operating costs of DSOs and MMGs increase accordingly, while the total social benefit decreases. This is because the system needs to reserve more standby capacity to cope with uncertainty, thus sacrificing some economic efficiency for higher reliability.

[0135] Table 2. Costs of DSO and MG with different robustness parameters

[0136]

[0137] To further quantify reliability, a "safety slack margin" was introduced as an evaluation index, which is defined as the minimum output value required to ensure the safe operation of the system without violating constraints. Figure 10 This paper presents the operating costs and safety relaxation margins (i.e., safety relaxation allowances) of MMGs under different robustness parameters. It can be seen that as the robustness parameters increase, the operating costs rise, but the safety relaxation margin decreases, indicating that the conservatism and operational reliability of the microgrid are enhanced. Therefore, in practical implementation, robustness parameters should be set reasonably according to requirements to achieve the optimal balance between economy and reliability.

[0138] To analyze the impact of peer-to-peer transactions on MG optimization scheduling, the scheduling results of Scheme 2 and Scheme 3 were compared. Figure 11 , 12 Table 13 shows the energy storage charging and discharging power of each MG and the power exchanged with the DSO. The results show that after introducing direct trading, the charging and discharging activities of each energy storage system are significantly reduced (because the MMGs achieve power mutual assistance through trading, reducing dependence on energy storage regulation). Furthermore, direct trading changes the interaction mode between MGs and the DSO: without direct trading (Scheme 3), MG1 and MG2 continuously sell electricity to the DSO, while MG3 continuously purchases electricity from the DSO; however, after introducing direct trading (Scheme 2), each MG prioritizes trading with other MGs, and the remaining power is then balanced with the DSO, resulting in a more balanced energy flow distribution. Table 3 compares the operating costs under the two schemes. The results show that after introducing direct trading, the operating costs of each MG tend to be balanced, and the total social benefit increases by 4.29%. Specifically, MG1 and MG2 obtain additional benefits by selling electricity to MG3, while MG3 meets its own needs at a lower cost than purchasing electricity from the DSO, achieving a win-win situation for all parties. In this specification, direct trading refers to point-to-point trading.

[0139] Table 3. Costs of DSOs and MGs with and without direct energy trading.

[0140]

[0141] This specific implementation case fully verifies the effectiveness of the method proposed in this invention. The robust optimization model effectively characterizes the risks brought about by uncertainty, the direct trading mechanism significantly improves the system's economy and flexibility, and the solution algorithm based on KKT conditions and strong duality theory ensures efficient solution of the model. The technology described in this specification provides a comprehensive technical solution for MMGs with a high proportion of renewable energy to participate in market transactions.

[0142] Please see Figure 14 The diagram shown is a structural schematic of an electronic device provided in an embodiment of this specification.

[0143] like Figure 14 As shown, the electronic device 1100 may include: at least one processor 1101, at least one network interface 1104, a user interface 1103, a memory 1105, and at least one communication bus 1102. The communication bus 1102 can be used to connect and communicate with the various components mentioned above. The user interface 1103 may include buttons, and optionally may include standard wired or wireless interfaces. The network interface 1104 may include, but is not limited to, a Bluetooth module, an NFC module, or a Wi-Fi module. The processor 1101 may include one or more processing cores. The processor 1101 connects to various parts within the electronic device 1100 using various interfaces and lines, and performs various functions of the routing device and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 1105, and by calling data stored in the memory 1105. Optionally, the processor 1101 may be implemented using at least one hardware form of DSP, FPGA, or PLA. The processor 1101 may integrate one or more combinations of CPU, GPU, and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content that the display screen needs to show; and the modem is used for wireless communication.

[0144] It is understandable that the aforementioned modem may not be integrated into the processor 1101, but may be implemented using a separate chip.

[0145] The memory 1105 may include RAM or ROM. Optionally, the memory 1105 may include a non-transitory computer-readable medium. The memory 1105 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 1105 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 1105 may also be at least one storage device located remotely from the aforementioned processor 1101. As a computer storage medium, the memory 1105 may include an operating system, a network communication module, a user interface module, and application programs. The processor 1101 may be used to call the application programs stored in the memory 1105 and execute the methods in the above-described embodiments.

[0146] This specification also provides a computer-readable storage medium storing instructions that, when executed on a computer or processor, cause the computer or processor to perform multiple steps as described in the above embodiments. If the constituent modules of the above-described electronic device are implemented as software functional units and sold or used as independent products, they can be stored in the computer-readable storage medium.

[0147] This specification also provides a computer program product, including a computer program that, when executed by a processor, implements the multiple steps described in the above embodiments.

[0148] Where there is no conflict, the technical features in this embodiment and implementation scheme can be combined arbitrarily.

[0149] In the above embodiments, the implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. Those skilled in the art should also understand that by simply performing some logic programming on the method flow using the aforementioned hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0150] The embodiments described above are merely preferred embodiments of this specification and are not intended to limit the scope of this specification. Any modifications and improvements made by those skilled in the art to the technical solutions of this specification without departing from the spirit of this specification should fall within the protection scope defined by the claims of this specification.

Claims

1. A robust optimization scheduling method for multi-microgrid participation in the electric vehicle auxiliary market, characterized in that, Including the following steps: A three-layer dispatch architecture is constructed, consisting of an upper-layer auxiliary power market clearing layer, a middle-layer active distribution network layer, and a lower-layer multi-microgrid layer. The active distribution network layer acts as an agent for multiple microgrids in the lower layer to bid and clear in the auxiliary power market with the upper-level power grid, and converts the market results into incentive prices and distributes them to each microgrid. A multi-market clearing model is established based on the supply-side price quota curve and the demand-side price quota curve; The active distribution network layer selects either the supply-side price quota curve or the demand-side price quota curve to participate in the bidding for the auxiliary power market based on the net load status of the microgrid in the next time period t, and obtains the market results, which include the cleared power volume and the cleared power price. A distribution optimization scheduling model for the active distribution network layer is established, with the objective of minimizing the total operating cost of the distribution network system. Optimization scheduling models for each microgrid are established, with the goal of minimizing the operating cost of the microgrid. Each microgrid generates its own real-time scheduling strategy based on its optimization scheduling model, which is then executed by the microgrid in the next time period t. Methods for establishing multi-market clearing models based on supply-side price quota curves and demand-side price quota curves include: The supply-side price quota curve and the demand-side price quota curve are respectively converted into several discrete price segments, and each price segment records the bid price and the corresponding maximum tradable electricity volume. A status flag is introduced to indicate whether the supply-side price quota curve and the price quota curve segment of the demand-side price quota curve are fully referenced in the market clearing model solution; For a quoted price segment that is referenced but not fully referenced, a variable with continuous values ​​is set to represent the electricity volume of the quoted price segment referenced by the market clearing model, which is denoted as the transaction electricity volume. The market-clearing price is determined by the intersection of total market supply and total demand, and the estimated bidding strategy is obtained based on the historical bidding strategies of the active distribution network layer. A market clearing model is constructed, wherein the input of the market clearing model is the estimated bidding strategy of the active distribution network layer, and the output is the clearing price and the clearing volume; The active distribution network layer selects either the supply-side price quota curve or the demand-side price quota curve to participate in the bidding for the auxiliary power market based on the net load status of the microgrid in the next time period t. Methods for obtaining market results include: Read the predicted load and predicted power generation of each microgrid in the next time period t, calculate the net load of each microgrid, determine the market role of each microgrid based on the net load of each microgrid, and then select the supply-side price quota curve or the demand-side price quota curve, and input it into the market clearing model. Based on the transformed supply-side price quota curve or demand-side price quota curve, and the estimated bidding strategy, the market clearing model is solved to obtain the intersection point, and the market result is obtained based on the intersection point. The optimal scheduling model for microgrids is a two-stage robust optimization model, which includes a first-stage robust optimization model and a second-stage robust optimization model. The first-stage robust optimization model reads the day-ahead predicted load and day-ahead predicted generation output for each time period, calculates the day-ahead net load for each microgrid for each time period, solves the market clearing model based on the day-ahead net load, obtains the day-ahead market results, and generates the day-ahead dispatch strategy for the microgrid based on the day-ahead market results. The second-stage robust optimization model reads the predicted load and predicted power generation for the next time period t, calculates the net load for each microgrid in the next time period t, solves the market clearing model based on the net load, obtains the market results, and generates a real-time dispatch strategy for the microgrid based on the market results.

2. The robust optimization scheduling method for multi-microgrid participation in the electric vehicle auxiliary market according to claim 1, characterized in that, While receiving incentive tariffs from the active distribution network layer, microgrids also engage in point-to-point electricity trading with other microgrids. The total operating cost of the power distribution network system includes the operation and maintenance costs of micro gas turbines, the operation and maintenance costs of renewable energy, the charging, discharging and driving costs of electric vehicle clusters, the revenue from energy trading with the microgrid, market clearing fees, and carbon trading costs. The constraints of the power distribution optimization scheduling model include power flow constraints of the power distribution network, upper and lower limits of distributed power output, electric vehicle charging and discharging logic and energy state constraints, upper and lower limits of incentive electricity prices, and carbon emission quota and actual emission calculation constraints.

3. A robust optimization scheduling method for multi-microgrid participation in the electric vehicle auxiliary market according to claim 1, characterized in that, While receiving incentive tariffs from the active distribution network layer, microgrids also engage in point-to-point electricity trading with other microgrids. The operating cost of the microgrid's optimal scheduling model includes the operation and maintenance costs of the energy storage system, the energy interaction costs with the active distribution network layer, and the point-to-point transaction costs. The constraints of the microgrid's optimal scheduling model include active power balance constraints, reactive power balance constraints, distributed power output constraints, energy storage system charging and discharging power and energy state constraints, point-to-point bilateral trading balance constraints, and carbon trading constraints.

4. A robust optimization scheduling system for multi-microgrid participation in the electric vehicle auxiliary power market, characterized in that, include: The architecture module constructs a three-layer scheduling architecture, including an upper-layer auxiliary power market clearing layer, a middle-layer active distribution network layer, and a lower-layer multi-microgrid layer. The active distribution network layer acts as an agent for multiple microgrids in the lower layer to bid and clear in the auxiliary power market with the upper-level power grid, and converts the market results into incentive prices and distributes them to each microgrid. The model module establishes a multi-market clearing model based on the supply-side price quota curve and the demand-side price quota curve; The bidding module allows the active distribution network layer to select either the supply-side price quota curve or the demand-side price quota curve to participate in the bidding for the auxiliary power market based on the net load status of the microgrid in the next time period t, and obtain the market results, which include the cleared power volume and the cleared power price. The first optimization module establishes a distribution optimization scheduling model for the active distribution network layer, with the goal of minimizing the total operating cost of the distribution network system. The second optimization module establishes an optimization scheduling model for each microgrid, with the goal of minimizing the operating cost of the microgrid. The scheduling module generates real-time scheduling strategies for each microgrid based on its own optimized scheduling model, which are then executed by the microgrid in the next time period t. Methods for establishing multi-market clearing models based on supply-side price quota curves and demand-side price quota curves include: The supply-side price quota curve and the demand-side price quota curve are respectively converted into several discrete price segments, and each price segment records the bid price and the corresponding maximum tradable electricity volume. A status flag is introduced to indicate whether the supply-side price quota curve and the price quota curve segment of the demand-side price quota curve are fully referenced in the market clearing model solution; For a quoted price segment that is referenced but not fully referenced, a variable with continuous values ​​is set to represent the electricity volume of the quoted price segment referenced by the market clearing model, which is denoted as the transaction electricity volume. The market-clearing price is determined by the intersection of total market supply and total demand, and the estimated bidding strategy is obtained based on the historical bidding strategies of the active distribution network layer. A market clearing model is constructed, wherein the input of the market clearing model is the estimated bidding strategy of the active distribution network layer, and the output is the clearing price and the clearing volume; The active distribution network layer selects either the supply-side price quota curve or the demand-side price quota curve to participate in the bidding for the auxiliary power market based on the net load status of the microgrid in the next time period t. Methods for obtaining market results include: Read the predicted load and predicted power generation of each microgrid in the next time period t, calculate the net load of each microgrid, determine the market role of each microgrid based on the net load of each microgrid, and then select the supply-side price quota curve or the demand-side price quota curve, and input it into the market clearing model. Based on the transformed supply-side price quota curve or demand-side price quota curve, and the estimated bidding strategy, the market clearing model is solved to obtain the intersection point, and the market result is obtained based on the intersection point. The optimal scheduling model for microgrids is a two-stage robust optimization model, which includes a first-stage robust optimization model and a second-stage robust optimization model. The first-stage robust optimization model reads the day-ahead predicted load and day-ahead predicted generation output for each time period, calculates the day-ahead net load for each microgrid for each time period, solves the market clearing model based on the day-ahead net load, obtains the day-ahead market results, and generates the day-ahead dispatch strategy for the microgrid based on the day-ahead market results. The second-stage robust optimization model reads the predicted load and predicted power generation for the next time period t, calculates the net load for each microgrid in the next time period t, solves the market clearing model based on the net load, obtains the market results, and generates a real-time dispatch strategy for the microgrid based on the market results.

5. An electronic device, characterized in that, Including the processor and memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code stored in the memory to perform the method as described in any one of claims 1-3.

6. 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 method as described in any one of claims 1-3.

7. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-3.

Images