Scene-based random optimization method for micro-grid energy and flexibility scheduling

Abstract

The invention provides a scene-based random optimization method for micro-grid energy and flexibility scheduling, and belongs to the technical field of micro-grid optimization scheduling. The method comprises the following specific steps: minimizing the expected cost of the micro-grid as a target; setting a power balance constraint in the micro-grid according to a relation among photovoltaic power generation capacity, power load power consumption, battery energy storage charging and discharging power and power loss in the micro-grid internal transmission and conversion process in the micro-grid; determining a flexible service constraint of the micro-grid according to the flexible service model and the battery energy storage model; predicting and updating by using a rolling time domain method, modeling uncertainty of the micro-grid by combining scene generation and scene reduction technologies, determining a day-ahead prediction error of the micro-grid, and constructing a random optimization model of energy and flexibility of the micro-grid; based on a flexible intra-day framework, solving a random optimization model of micro-grid energy and flexibility according to a scene at different time stages, and obtaining a flexible scheduling strategy; meanwhile, the stability and efficiency of the micro-grid are considered.

Description

Technical Field

[0001] The present invention relates to the technical field of microgrid optimal scheduling, and particularly to a stochastic optimization method for microgrid energy and flexibility scheduling based on scenarios. Background Art

[0002] With the continuous increase in the penetration rate of renewable energy in the power system, the system's demand for flexibility is also increasing day by day. Distributed energy resources and demand response resources can provide demand-side flexibility services, and fixed battery energy storage systems have attracted much attention due to their fast charge and discharge capabilities. In a microgrid, the battery energy storage system can not only store energy but also provide flexibility services in the grid-connected mode, which helps to balance supply and demand and reduce costs. Microgrid flexibility scheduling aims to optimize energy utilization and respond to changes while ensuring system stability. Without disrupting the component structure of the microgrid and the power transmission and distribution process, it dynamically controls and flexibly adjusts the operating state. This method is valuable in various microgrid scenarios, can reduce costs, reduce risks, improve reliability, economy, and environmental friendliness, adapt to various conditions and requirements, and is playing an increasingly important role in the future energy pattern.

[0003] However, existing microgrid flexibility scheduling methods only focus on estimating technically feasible flexibility, do not fully consider the economic willingness of flexibility providers, lack quantification and evaluation of flexibility, especially at the distribution level.

[0004] Therefore, there is an urgent need for a microgrid energy and flexibility scheduling method that can comprehensively consider uncertainty, market strategies, and capacity constraints. Summary of the Invention

[0005] In view of this, the present invention provides a stochastic optimization method for microgrid energy and flexibility scheduling based on scenarios. By constructing a scenario-based stochastic optimization model, an energy and flexibility scheduling strategy for a microgrid including photovoltaic power generation and a fixed battery energy storage system is obtained to minimize the operating cost of the microgrid and maximize the flexibility benefit.

[0006] To this end, the present invention provides the following technical solutions:

[0007] A stochastic optimization method for microgrid energy and flexibility scheduling based on scenarios, comprising:

[0008] Taking minimizing the expected cost of the microgrid as the objective;

[0009] According to the relationship between the photovoltaic power generation amount, the electricity consumption of the power load, the charge and discharge power of the battery energy storage, and the power loss in the internal transmission and conversion process of the microgrid, setting the power balance constraint in the microgrid;

[0010] Determine the flexibility service constraints of the microgrid according to the flexible service model and the battery energy storage model;

[0011] Use the rolling horizon method for prediction update, combine the scenario generation and scenario reduction techniques to model the uncertainty of the microgrid, determine the day-ahead prediction error of the microgrid, and construct a stochastic optimization model for the energy and flexibility of the microgrid;

[0012] Based on the intra-day framework of flexibility, solve the stochastic optimization model of the microgrid energy and flexibility according to the scenarios at different time stages to obtain the flexibility scheduling strategy.

[0013] Furthermore, the expected cost of the microgrid includes:

[0014] The expected cost of environmental protection, the expected cost of input energy, the expected revenue of output energy, the expected revenue of providing flexibility services, the expected cost of obtaining peak power from the main grid, and the expected cost of battery energy storage.

[0015] Furthermore, the microgrid balancing power constraint includes:

[0016]

[0017] Among them, and are the photovoltaic power generation and the electricity consumption of the power load respectively; and are the charging and discharging powers of the battery energy storage respectively; is the power loss during the internal transmission and conversion process of the microgrid.

[0018] Furthermore, the flexible service model includes:

[0019]

[0020] Among them, the variable is the net power peak of scenario w during the flexibility activation period ; when the net power peak of the microgrid does not exceed the variable p fl,p , it means that the flexibility service is realized.

[0021] Furthermore, the battery energy storage model includes:

[0022] The state of charge of the battery energy storage:

[0023]

[0024] The capacity loss of the battery energy storage caused by cyclic aging:

[0025]

[0026] Expected cost of battery energy storage:

[0027]

[0028] wherein, SoE t is the state of energy of the battery energy storage; SoE init is the initial state of energy of the battery energy storage; E max represents the capacity of the battery energy storage at the start of the scheduling period; μ ch and μ dis are the charging and discharging efficiencies of the battery energy storage respectively; q is the battery capacity loss; B1 and B2 are the cycle aging coefficients; I C is the average charging rate; C B,0 is the battery procurement cost; H is the remaining capacity at the end of the battery life.

[0029] Furthermore, the modeling of the microgrid uncertainty includes:

[0030] Using Gaussian distribution to simulate the day-ahead prediction errors of photovoltaic power generation and power consumption in the microgrid;

[0031] Using the Monte Carlo method to generate a number of scenarios, and generating representative scenarios through scenario reduction techniques.

[0032] Furthermore, the stochastic optimization model of the microgrid energy and flexibility includes:

[0033]

[0034] f' = c im - r ex + c p + c B + c e

[0035]

[0036] wherein, f flex represents the expected cost of the microgrid providing flexibility services; f' represents the expected cost of the microgrid when no benefit is obtained from the flexibility services; f" represents the expected cost of the microgrid when no microgrid services are provided; t0 represents the flexibility request notification time; t2 represents the end time of the flexibility activation period; β t,w is the bid for deviation from flexibility, i.e., the flexibility deviation amount at scenario w and time t; C pen is the penalty for not providing flexibility services, i.e., the unit deviation cost.

[0037] Furthermore, the flexibility-based intraday framework solves the stochastic optimization model of the microgrid energy and flexibility according to scenarios at different time stages to obtain the flexibility scheduling strategy, including:

[0038] Before receiving the notification of the flexibility service request, solve the microgrid control strategy with the goal of minimizing the operating cost of the microgrid;

[0039] After receiving the flexibility service request, calculate the flexibility capacity and evaluate the flexibility; during the activation period of the flexibility service, solve the microgrid control strategy with the goal of meeting the flexibility demand and minimizing the deviation cost.

[0040] Advantages and positive effects of the present invention:

[0041] The scenario-based stochastic optimization model proposed by the present invention can effectively determine the energy and flexibility scheduling scheme of the microgrid while considering the uncertainties of photovoltaic power generation and power load demand. By quantifying the flexibility of the microgrid, considering uncertainties, market energy scheduling strategies and capacity-limited flexibility service models, it provides a more accurate and cost-effective flexibility evaluation and scheduling method. Combined with the rolling horizon method, the set points can be dynamically updated to reduce the impact of prediction errors, ensure that the flexibility is scheduled according to the bidding requirements, and improve the economic value of the microgrid.

[0042] The expected cost of the microgrid considering environmental protection includes the expected cost of carbon emissions and the expected revenue of green subsidies. When making energy and flexibility scheduling decisions, the microgrid can more comprehensively consider the impact of environmental factors on costs and revenues, which helps to promote the microgrid to develop in a more environmentally friendly and sustainable direction.

[0043] The flexibility service model provides strategies for microgrid operators to trade in the energy market and the flexibility market, enabling them to flexibly adjust energy and flexibility supply according to market demand and system status, achieve optimal resource allocation, reduce costs for microgrid customers, and at the same time provide demand-side flexibility services for grid operators, relieve the pressure of peak demand on the grid, and improve the economy and sustainability of grid operation.

[0044] This method is based on a one-day framework, dynamically updates the uncertainty model, considers various factors, can be used for flexibility supplier evaluation and bidding strategy design, and also helps system operators design incentive mechanisms, can improve the economic value of the microgrid and enhance its ability to cope with uncertainties.

[0045] Adopt the rolling horizon method, only implement the optimal control decision of the next time step, then update the prediction information and re-solve the stochastic optimization problem. By continuously repeating this process, real-time control of the microgrid is achieved and the impact of prediction errors is reduced.

[0046] The scenario reduction technique, the obtained scenarios represent the true variability of the input values, while hardly compromising the accuracy of the results. Through the scenario reduction technique, each scenario has a different occurrence probability, which improves the solution efficiency while ensuring accuracy.

[0047] The method of the present invention can be applied to different case studies, such as different notification times or flexibility activation periods, with strong versatility and flexibility, which helps microgrid operators evaluate flexibility potential and value, and can also provide reference for grid operators to formulate flexibility prices and incentives, promoting the optimization of grid capacity reserves. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0049] Figure 1 It is a flowchart of a stochastic optimization method for scenario-based microgrid energy and flexibility scheduling in an embodiment of the present invention;

[0050] Figure 2 It is a diagram of the flow direction of microgrid resources and energy power in an embodiment of the present invention;

[0051] Figure 3 It is a flowchart of microgrid energy and flexibility scheduling by the rolling horizon method in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] In order to enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0053] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order different from those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0054] The present invention provides a stochastic optimization method for scenario-based microgrid energy and flexibility scheduling. Aiming at the uncertainties of renewable energy generation and power demand in the microgrid, a stochastic optimization model is constructed by considering the power losses in the internal transmission and conversion processes of the microgrid; and with the goal of minimizing the expected cost (energy cost, peak power cost, battery aging cost, environmental protection cost and flexibility income) of the microgrid providing flexibility services, a new power balance formula is introduced to precisely coordinate the internal energy flow of the microgrid; by constructing a flexibility service model including a battery energy storage model and a capacity limit service model, combined with the rolling horizon method to dynamically adjust the scheduling strategy, and modeling the uncertainties based on scenario generation and reduction technology, the stochastic optimization problem is solved according to the situation in different time stages to achieve flexibility evaluation and scheduling.

[0055] Combined with Figure 1 The method of the present invention is further described as follows:

[0056] S101. Taking the minimization of the expected cost of the flexibility services that the microgrid can provide as the optimization goal, a stochastic optimization model for the energy and flexibility problems of the microgrid is constructed; in the embodiment of the present invention, the microgrid uses a photovoltaic system and a stationary battery energy storage as flexibility resources.

[0057] Specifically, the goal of the scenario-based stochastic optimization model is to consider the uncertainties of power load demand and photovoltaic power generation, and minimize the operating cost of the microgrid in a future period of time.

[0058] Predict the statistical distributions of photovoltaic power generation, load and prediction errors, and use them as the input of the comprehensive stochastic optimization model. A reasonable selection of the time period length avoids the need to predict the electricity price.

[0059] The cost of the microgrid includes the energy cost (which can be negative because the income from selling energy is subtracted), the peak power cost, the battery energy storage degradation cost, the expected cost considering environmental protection, and the income from providing flexibility.

[0060] In this embodiment, the optimization goals of the stochastic optimization model include:

[0061] The microgrid optimization goal, and the objective function is to minimize the expected cost of the microgrid that can provide flexibility services during the scheduling period;

[0062] The objective function in this embodiment is:

[0063] minf flex =c im -r ex +c p +c B +c e -r flex

[0064] Expected cost of input energy:

[0065]

[0066] Expected revenue of output energy:

[0067]

[0068] Expected cost of obtaining peak power from the main grid:

[0069]

[0070] Expected revenue of providing flexibility services:

[0071]

[0072] where minf fl ex is the objective function to minimize the expected cost of the microgrid that can provide flexibility services; c im and r ex are the expected costs of input energy and output energy respectively; c p is the expected cost of obtaining peak power from the main grid; c B is the expected cost of battery energy storage; c e is the expected cost considering environmental protection; r flex is the expected revenue of providing flexibility services; H and Δt represent the scheduling level and the time interval length respectively; the positive variables and are the input and output powers of the transmission grid at time step t and scenario w respectively; ∏ w is the probability of scenario w occurring; Λ t represents the real-time electricity price; C i is the transmission grid charge of the power grid, that is, the power grid utilization rate; C e is the compensation cost paid by the distribution system operator to incentivize the reduction of network losses; p flex represents the capacity of active power flexibility; C flex is the flexibility price; is the original peak power of the microgrid, p fl,p is the expected peak input power during the flexibility activation period.

[0073] Introducing the expected cost c e considering environmental protection into the expected cost of the microgrid is closer to the actual operating cost of the microgrid.

[0074] Flexibility in the microgrid is calculated based on the reduction of power capacity. When flexibility is activated, the microgrid resources modify their schedules to ensure that the peak input power does not exceed the new capacity provided by the flexibility service. Parameter p flex The peak value should be based on the value agreed upon by the distribution system operator and the microgrid operator.

[0075] In this embodiment, the microgrid considers the expected costs of environmental protection, including:

[0076] Expected costs of microgrid environmental protection:

[0077]

[0078] Expected costs of carbon emissions:

[0079]

[0080] Expected benefits of green subsidies:

[0081]

[0082] Among them, and are the expected costs of carbon emissions and the expected benefits of green subsidies respectively; represents the carbon emission cost coefficient of the input power; represents the carbon emission cost coefficient of battery charging; is the carbon emission cost coefficient of battery discharge; represents the subsidy coefficient of photovoltaic power generation; represents the subsidy coefficient of flexibility service.

[0083] S102. According to the relationships among variables such as the photovoltaic power generation amount, the electricity consumption of the power load, the charge and discharge power of the battery energy storage, and the power loss in the internal transmission and conversion process of the microgrid, set the power balance in the microgrid; construct the battery energy storage model and the flexibility service model of the microgrid.

[0084] 1) Power balance of the microgrid:

[0085]

[0086] Among them, and are the photovoltaic power generation amount and the electricity consumption of the power load respectively; and are the charge and discharge power of the battery energy storage respectively; is the power loss in the internal transmission and conversion process of the microgrid.

[0087] In the power balance formula of the microgrid, the power losses during the internal transmission and conversion processes in the microgrid are introduced. Accurately considering the power losses enables better coordination of the internal energy flow within the microgrid during power exchange, which helps in better planning and management of the power system.

[0088] 2) The converter that couples the battery energy storage and the photovoltaic system operates bidirectionally. Therefore, both the power generation of the photovoltaic and the power discharged from the battery energy storage can meet the power load demand in the microgrid and be output to the AC grid. And the battery energy storage can be charged through the upstream AC grid and the photovoltaic system.

[0089] As Figure 2 shown, the microgrid 200 in this embodiment includes:

[0090] 201 is the power load; 203 is the AC grid; 204 is the AC / DC conversion converter; 202 is the coupling among 201, 203, and 204. 205 is the battery energy storage; 206 is the photovoltaic power generation; 207 is the coupling among 204, 205, and 207.

[0091] The arrow direction represents the power flow direction: the power flow directions of the power load and the photovoltaic power generation are unidirectional, while the power flow directions of the AC grid and the battery energy storage are bidirectional. The AC / DC conversion converter serves as the bridge for energy exchange, connecting each system to complete energy interconversion and maintaining the power balance of the microgrid.

[0092] In this embodiment, a battery energy storage model of the microgrid is constructed:

[0093] The energy state of the battery energy storage:

[0094]

[0095] The capacity loss of the battery energy storage caused by cyclic aging:

[0096]

[0097] The expected cost of the battery energy storage:

[0098]

[0099] Among them, soe t is the energy state of the battery energy storage; SoE init is the initial energy state of the battery energy storage; E max represents the capacity of the battery energy storage at the start of the scheduling period; μ ch and μ dis are the charging and discharging efficiencies (percentage) of the battery energy storage respectively; q is the battery capacity loss (percentage); B1 and B2 are the cyclic aging coefficients; I Cis the average charging rate; C B,0 is the battery procurement cost; H is the remaining capacity (percentage) at the end of the battery life.

[0100] The battery energy storage model constructed in this embodiment takes into account factors such as charging efficiency, discharging efficiency, charging power, discharging power, and time interval, and can accurately describe the change of the energy state of the battery at different times; it helps to accurately grasp the remaining power and power supply time of the battery, which is crucial for the energy management of the microgrid. According to the current state of the battery and future load demand, the battery energy storage model reasonably arranges the charging and discharging operations of the battery to achieve optimal energy scheduling.

[0101] 3) In this embodiment, the flexibility is evaluated through the flexibility service model of the microgrid:

[0102]

[0103] where the variable is the flexibility activation period is the net power peak of scenario w at time. When the microgrid operator activates the flexibility, the dispatching of the battery energy storage power is targeted at meeting the flexibility capacity p flex If the net power peak of the microgrid does not exceed the variable p fl,p , the flexibility service can be realized.

[0104] S103. Use the rolling horizon method to dynamically adjust the dispatching strategy, dynamically update the prediction information to achieve real-time control, and model the uncertainty based on the scenario generation and reduction technology. In the operation and management of the microgrid, use the rolling horizon method to dynamically predict and update and closely cooperate with the scenario generation and reduction technology to cope with uncertainty and achieve efficient dispatching.

[0105] 1) Under closed-loop control, the rolling horizon method obtains accurate battery energy storage set points by repeatedly solving optimization problems, implementing the next optimal set point at each time step and shifting the dispatching level; the dynamic update of the prediction information adjusts its length and updates the scenario according to the latest predictions of the load and photovoltaic power output, providing a key basis for dispatching and adapting to the fluctuations of the load and generation;

[0106] 2) Based on the scenario generation and reduction technology, assume that the prediction error follows a specific Gaussian distribution and adjust the standard deviation. After generating scenarios using the Monte Carlo method, use the general algebraic modeling system to reduce the scenarios, providing effective inputs for the model, enhancing the ability to handle uncertainty, and ensuring the stable operation and optimal dispatching of the microgrid.

[0107] Specifically, the scenario-based stochastic optimization model can solve open-loop stochastic optimization problems and also solve closed-loop, i.e., rolling horizon, stochastic optimization problems. In closed-loop control, first, the optimal setpoint for the next time step is achieved; then, the scheduling horizon is shifted and the same problem is solved again (which was only solved once in open-loop control). The closed-loop control repeatedly solves the same problem multiple times (depending on the choice of time discretization steps) to obtain the battery energy storage setpoint for the same time period as in open-loop control; after the time horizon is shifted, before the next simulation, the prediction profile will be updated to account for the most recently available predictions and include the part of the time horizon not considered in the previous simulation.

[0108] To represent the uncertainty of the input values, it is assumed that the day-ahead prediction errors of photovoltaic power generation and electricity consumption respectively follow Gaussian distributions, which are N(0, 0.1 2 ) and N(0, 0.05 2 ). The standard deviations are used for the day-ahead prediction error distributions For the time step of the previous hour, and the time steps after the previous hour and up to six hours before, they are respectively equal to and Based on the above distributions, the Monte Carlo method is used to generate a number of scenarios, that is, randomly sample the input variables (taken from their most recently updated predictions) and add noise to represent the prediction errors;

[0109] The error generated by using a Gaussian random number generator for each time step of the available load or photovoltaic power generation prediction curve is then used to adjust the values of the base scenario to obtain a future scenario; the generated error is used to adjust the values of the base scenario, and finally an event scenario of future power load and photovoltaic power generation is obtained; then this process is repeated to generate all scenarios;

[0110] After generating the scenarios, the fast backward or forward mixing method in the SCENRED tool in GAMS is used to create a smaller number of scenarios (non-equivalent scenarios) that can represent the actual variations of the input values. Different occurrence probabilities are assigned to each scenario through scenario reduction techniques.

[0111] S104. Corresponding to the intra-day framework for obtaining flexibility, the stochastic optimization problem is solved according to the situation at different time stages to achieve flexibility assessment and scheduling, including:

[0112] The stochastic optimization problem is solved according to the situation at different time stages to achieve flexibility assessment and scheduling. The optimal strategy is determined by solving the stochastic optimization problem for each time step according to different situations. The operating cost is optimized before the flexibility request notification. After receiving the request, the flexibility capacity is calculated and the flexibility is evaluated. During the activation period, the strategy is adjusted to meet the demand and the deviation cost is minimized.

[0113] In this embodiment, the stochastic optimization problem is solved according to the situation in different time stages to achieve flexibility assessment and scheduling, including:

[0114] At each time step, the stochastic optimization problem is solved according to different situations to determine the optimal energy scheduling and flexibility provision strategy of the microgrid. Before the flexibility request notification, the optimization goal is to minimize the operating cost of the microgrid; after receiving the request, the flexibility capacity is calculated and the flexibility is evaluated; during the flexibility activation period, the control strategy is adjusted to meet the flexibility demand and minimize the deviation cost β t,w C pen Obtain the one-day framework for flexibility. Solve once at each time step τ to solve the following stochastic optimization problem, and the calculation formula is:

[0115]

[0116] f' = c im -r ex +c p +c B +c e

[0117]

[0118] where f flex represents the objective function for obtaining benefits from flexibility services; f' represents the objective function for not obtaining benefits from flexibility services; f" represents the objective function for the penalty for not providing flexibility services; t0 and t2 are the flexibility request notification time and the end time of the flexibility activation period respectively; β t,w is the bid for deviation from flexibility, that is, the flexibility deviation amount under scenario w and time t; C pen is the penalty for not providing flexibility services, that is, the unit deviation cost.

[0119] The microgrid calculates the optimal energy scheduling obtained by solving the stochastic optimization problem given by min f at each time step τ where τ < t0; if the distribution system operator needs to purchase flexibility services, a request is sent to the microgrid energy management system at τ = t0, and the microgrid energy management system randomly evaluates the flexibility that can be provided at τ = t0 by solving min f and responds with the flexibility amount. Before the next solution, the microgrid energy management system also receives a notification regarding the acceptance or rejection of p flex . If accepted, the microgrid solves min f when t0 < τ < t2, that is, until the solution boundary moves out of the activation period.

[0120] The goal of the optimization model is to minimize the expected energy, peak power cost, environmental protection cost, and battery aging cost, while maximizing the expected flexibility benefit. Considering the latest predicted curves of photovoltaic power generation and power demand, the uncertainty model is dynamically updated to solve the stochastic optimization problem under the rolling horizon. This model can be used by flexibility providers to evaluate their flexibility and design bidding strategies, and also by system operators to design incentive mechanisms for flexibility providers.

[0121] Combined with Figure 3 As shown, the process of microgrid energy and flexibility scheduling using the rolling horizon method in the embodiments of the present invention includes:

[0122] 1) First, perform initialization operations, set the initial input data, including the active power of photovoltaic power generation payload the probability ∏ of the scenario occurrence w , time t, the real-time electricity price Λ t and the initial energy state SOE of the battery init .

[0123] 2) Start the iteration. In each iteration:

[0124] The input data is based on the scenarios w = 1, 2, K, W and time t = 1, 2, K, T.

[0125] According to whether benefits are obtained from flexibility services, enter different branches of the flexibility service model; specifically including: no benefits obtained from flexibility services, benefits obtained from flexibility services, and penalties for not providing flexibility services.

[0126] Solve the scenario-based stochastic optimization problem through the battery energy storage model and the power balance formula.

[0127] 3) Continue the iteration until the calculations for all time periods and scenarios are completed, and output the data.

[0128] The method of the present invention can effectively quantify and optimize the flexibility of the microgrid, enhance the economic value, be applicable to various scenarios, provide decision-making support for microgrid operators, contribute to the optimization of the grid capacity reserve, and has high application value in the field of microgrid optimal scheduling.

[0129] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A scenario-based stochastic optimization method for microgrid energy and flexibility scheduling, characterized in that: include: The goal is to minimize the expected cost of the microgrid; According to the relationship between the photovoltaic power generation, power load power consumption, battery energy storage charging and discharging power, and power loss in the transmission and conversion process within the microgrid, the power balance constraint in the microgrid is set; Determine the flexibility service constraints of the microgrid based on the flexibility service model and battery energy storage model; The prediction update is made by using the rolling horizon method, and the scenario generation and scenario reduction techniques are combined to model the uncertainty of the microgrid, determine the day-ahead prediction error of the microgrid, and construct a stochastic optimization model of the energy and flexibility of the microgrid; Based on the intraday framework of flexibility, the stochastic optimization model of the energy and flexibility of the microgrid is solved according to the scenarios at different time stages to obtain the flexibility scheduling strategy.

2. A scenario-based stochastic optimization method for microgrid energy and flexibility scheduling according to claim 1, characterized in that: The expected costs of the microgrid include: The expected cost of environmental protection, the expected cost of imported energy, the expected benefits of exported energy, the expected benefits of providing flexibility services, the expected cost of taking peak power from the main grid, and the expected cost of battery storage.

3. The method for stochastic optimization of scenario-based microgrid energy and flexibility scheduling according to claim 1, characterized in that: The microgrid balancing power constraint includes: in, and They are photovoltaic power generation and power load power consumption respectively; and are the charging and discharging power of the battery energy storage, respectively; is the power loss during transmission and conversion within the microgrid.

4. The method for stochastic optimization of scenario-based microgrid energy and flexibility scheduling according to claim 1, characterized in that: The flexible service model includes: Among them, the variable Flexibility activation period When the net power peak of scenario w does not exceed the variable p fl,p , indicating flexibility service implementation.

5. The method for stochastic optimization of scenario-based microgrid energy and flexibility scheduling according to claim 1, characterized in that: The battery energy storage model includes: Energy state of battery storage: Battery energy storage capacity loss caused by cycle aging: Expected costs of battery storage: Among them, soe t State of Energy for battery storage; SoE init is the initial energy state of the battery; E max Indicates the capacity of battery energy storage at the beginning of the scheduling period; μ ch and μ dis are the battery energy storage charging and discharging efficiencies respectively; q is the battery capacity loss; B1 and B2 are the cycle aging coefficients; I C is the average charging rate; C B,0 is the battery purchase cost; H is the retained capacity at the end of the battery life.

6. The method for stochastic optimization of scenario-based microgrid energy and flexibility scheduling according to claim 1, characterized in that: The microgrid uncertainty modeling includes: The day-ahead forecast error of photovoltaic power generation and power consumption in microgrids is simulated using Gaussian distribution. Several scenarios were generated using the Monte Carlo method, and representative scenarios were generated through scenario reduction techniques.

7. The method for stochastic optimization of scenario-based microgrid energy and flexibility scheduling according to claim 1, characterized in that: The stochastic optimization model of microgrid energy and flexibility includes: f'=c im -r ex +c p +c B +c e Among them, f flex represents the expected cost of the microgrid providing flexibility services; f' represents the expected cost of the microgrid when no benefits are obtained from flexibility services; f" represents the expected cost of the microgrid when no microgrid services are provided; t0 represents the flexibility request notification time; t2 represents the end time of the flexibility activation period; β t,w is the bid deviation from flexibility, i.e., the flexibility deviation under scenario w and time t; C pen The penalty for not providing flexibility is the unit deviation cost.

8. A scenario-based stochastic optimization method for microgrid energy and flexibility scheduling according to claim 7, characterized in that: The flexibility-based intraday framework solves the stochastic optimization model of the microgrid energy and flexibility according to the scenario at different time stages to obtain a flexibility scheduling strategy, including: Before receiving the flexibility service request notification, solving the microgrid control strategy with the goal of minimizing the microgrid operation cost; After receiving the flexibility service request, the flexibility capacity is calculated and the flexibility is evaluated; during the flexibility service activation period, the microgrid control strategy is solved with the goal of meeting the flexibility demand and minimizing the deviation cost.

Images