Optimized scheduling method and system for multi-main-body microgrid considering source-load uncertainty

By building a three-layer game model, optimizing the new energy power generation and electricity consumption, the problems of uncertainty and load fluctuations in new energy power generation in multi-subject microgrids have been solved, and more efficient energy utilization and system stability have been achieved, reducing the Energy waste and operating costs.

CN118983871APending Publication Date: 2024-11-19HUANENG LIAONING ENERGY SALES LLC
View PDF 0 Cites 5 Cited by

Patent Information

Application Number
CN202410882034.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-03
Publication Date
2024-11-19

AI Technical Summary

Technical Problem

The existing technology has shortcomings in dealing with the uncertainty of new energy power generation and the fluctuation of electricity load, which increases the complexity of the optimization scheduling of energy systems, especially in multi-subject microgrids, where the interactive relationship is complex and it is difficult to achieve high efficiency of the system. Optimized operation.

Method used

By obtaining parameter information and load demand data of each entity of the energy system, a three-layer game model is built, including the central scheduling layer, the aggregation scheduling layer and the energy management layer, the output of each energy is optimized, and the intermittent and randomness of new energy power generation are considered. , use energy storage technology to optimize the electricity price structure and output plan to achieve coordinated optimization of multiple entities.

Benefits of technology

It improves energy utilization efficiency, reduces energy waste, reduces dependence on traditional fossil energy, enhances the stability and resilience of the system, optimizes the overall performance and economic benefits of the system, and reduces the impact of price fluctuations on the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118983871A_ABST
    Figure CN118983871A_ABST
Patent Text Reader

Abstract

The invention discloses an optimal scheduling method and system for a multi-agent micro-grid considering source-load uncertainty, and the method comprises the steps: optimal scheduling of three levels: a central scheduling level takes a multi-source micro-grid operator-a comprehensive energy producer-a load aggregator as game subjects, and considers the electricity selling and supply-demand balance of an electricity market; the aggregation scheduling layer optimizes, schedules and manages the output of each benefit subject, so that the energy loss is minimum; according to the energy management level, through day-ahead-day two-stage multi-time scale rolling optimization, new energy power generation time sequence volatility and power utilization load uncertainty are processed, and a robust optimization method is adopted to process main power grid-multi-source micro-grid transaction volatility, so that the model can effectively balance demands of supply and demand parties; and meanwhile, efficient operation of the energy system is realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of energy system operation dispatch optimization, and in particular to an optimization dispatch method and system for a multi-agent microgrid taking source-load uncertainty into account. Background Art

[0002] With the global energy transformation and the "dual carbon" strategic goal, large-scale access to new energy sources such as wind and solar energy has become an inevitable trend. However, these energy sources have significant randomness and intermittency, which brings huge challenges to the traditional power system dispatch.

[0003] The integrated energy system achieves complementary and cascaded utilization of multiple energy sources by coupling multiple energy networks such as electricity, heat, and gas, thereby improving energy utilization efficiency and promoting the consumption of new energy.

[0004] The uncertainty of renewable energy generation, the fluctuation of electricity load, and the change of electricity trading prices have increased the complexity of optimizing the scheduling of energy systems. Existing technologies are still insufficient in dealing with these uncertainties and fluctuations.

[0005] The integrated energy system involves multiple stakeholders, such as microgrid operators, integrated energy producers and sellers, and load aggregators. The interactive relationships between these stakeholders are complex, which places higher demands on the optimized operation of the system. Summary of the invention

[0006] In view of the above-mentioned problems, the present invention is proposed.

[0007] Therefore, the technical problem solved by the present invention is: the problem of optimal scheduling of an integrated energy system under source-load uncertainty and price fluctuations.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions: an optimization scheduling method for a multi-agent microgrid taking into account source-load uncertainty, comprising:

[0009] Obtain parameter information of each energy subject in the energy system, user load demand data for energy, and sensitivity indicators of each stakeholder;

[0010] At the central scheduling layer, a total objective function is generated according to the sensitivity index and constrained to construct a first-layer decision model;

[0011] In the process of executing the total objective function, the output of each energy source is optimized through the aggregation scheduling layer, and the cost-sensitive index in operation is calculated to construct a cooperative game model involving multiple stakeholders in the second layer;

[0012] The energy management layer is used to analyze the wind power, photovoltaic new energy forecast curves and daily load curves to obtain the volatility of the sensitivity indicators when the multi-source microgrid purchases electricity from the main power grid. The overall objective function is fed back to construct the third-layer optimization scheduling model; through the three-layer game model, the optimization of the multi-agent integrated energy system is achieved.

[0013] As a preferred solution of the optimization scheduling method of the multi-agent microgrid taking into account the uncertainty of source-load described in the present invention, wherein: the energy subject includes three different types of energy subjects, namely wind power, photovoltaic power and energy storage power station; the parameter information includes the power information of charging and discharging of power plants and power stations;

[0014] The sensitivity index of each stakeholder includes a profit index generated when energy circulates between the stakeholder groups, which is used to represent the energy loss after value quantification, and the value of energy when it is obtained is represented by cost, and the value of energy when it is released is represented by price; in the optimization process, the minimum energy loss is achieved by maximizing the profit index;

[0015] The central dispatching layer receives information on abandoned wind and solar power and real-time transaction prices from the energy management layer, and based on the received information, constructs constraints and objective functions to optimize the electricity price structure with the goal of maximizing the total profit of the system.

[0016] As a preferred solution of the multi-agent microgrid optimization scheduling method taking into account source-load uncertainty described in the present invention, wherein: the objective function includes, the interest demands of the multi-agent integrated energy system are expressed as the maximization of the comprehensive profits of the three parks, the energy storage power station and the wind farm;

[0017] The objective mathematical model of the multi-agent integrated energy system is expressed as:

[0018]

[0019] Among them, F represents the total profit of the multi-agent integrated energy system; N represents the number of 3 parks in this paper; T represents the total scheduling time of 24 hours; C i is the operating profit of park i; C ess is the unit power profit of the energy storage power station; C grid For distribution network profits; is the user's demand response to energy e, where e represents the combination of electricity and thermal loads; is the price at which the park sells energy e to users at time t; is the amount of electricity purchased by park i from the distribution network at time t; represents the electricity selling price of the distribution network at time t; is the power loss of park i at time t; represents the unit profit of the power lost in park i at time t; is the amount of electricity delivered from park i to the energy storage at time t; represents the price of the electricity delivered by park i to the energy storage at time t; The amount of electricity sold by wind power to park i at time t; represents the electricity price of wind power at time t; is the power consumption of device B in park i at time t; is the unit power operation and maintenance cost of device B in park i, including operation cost and start-up and shutdown cost; represents the amount of electricity sold by park i to the distribution network at time t; represents the price of electricity sold by park i to the distribution network at time t; The economic compensation provided by the distribution network when responding to the demand of users in park i; is the amount of electricity delivered from the distribution network to the energy storage at time t; Represents the price of electricity delivered by the distribution network to the energy storage at time t; The amount of electricity sold by wind power to the distribution network at time t;

[0020] The demand response constraint is expressed as:

[0021]

[0022] Among them, P DR i,m,t P is the electric load power after the park responds to demand; L i,m,t and Q L i,m,t are the original electric load and heat load power of the community respectively; P cut i,m,t , Q cut i,m, t are the reduced power of the park’s electrical load and thermal load respectively; P tran i,m,t The power of the electric load shifted by the community in response to demand response. When the value is positive, it means that the electric load is transferred from other time periods to time period t, and vice versa; r cut is the maximum proportionality factor that can reduce the heat load.

[0023] As a preferred solution of the multi-agent microgrid optimization dispatching method taking into account source-load uncertainty described in the present invention, wherein: the aggregate dispatching layer obtains the electricity price information transmitted by the central dispatching layer and the demand-side response measures; according to the dispatching status and dispatching demands, the output plans of the wind farm, photovoltaic power station, energy storage power station, integrated energy producer and distributor, and energy conversion equipment are optimized to respond to changes in electricity prices and meet load demands;

[0024] On the premise of meeting the supply and demand balance, the production cost is minimized by adjusting the output ratio and time of each energy source. The optimization results of the aggregation scheduling layer are fed back to the central scheduling layer for the next electricity price optimization and evaluation of the total system profit.

[0025] As a preferred solution of the multi-agent microgrid optimization scheduling method taking into account source-load uncertainty described in the present invention, wherein: in the aggregate scheduling layer, the revenue model of the new energy electric field is:

[0026]

[0027] in, and They represent the charging power of the new energy electric field at time t, the power sold by the electric field to the distribution network, and the power of park i respectively; c w,o and W w They represent the unit price of electricity sold by new energy power plants, the unit power operation and maintenance cost and profit respectively; n is the value range of time, indicating the total scheduling time of 24 hours; N represents the number of 3 parks in this article;

[0028] Energy storage power station unit work profit model:

[0029]

[0030] in, Respectively represent the discharge and charging power of the energy storage power station to park i; P t w,w and P t c,w are the charging power obtained by the energy storage power station from the upper power grid and the centralized wind farm; c t e represents the comprehensive price of electricity sold by the energy storage power station; ψ and W z,s Respectively represent the unit power operation and maintenance cost and profit of the energy storage power station;

[0031] Constraints on purchasing electricity and heat:

[0032]

[0033] in, The power demand of community m for integrated energy operator i after completing interaction and demand response with other communities; The electrical load power after demand response for the community; represents the photovoltaic power generation of the mth park; if P m-n i,t >0, it means that the m park provides power to the n park. If P m-n i,t <0, it means that park n provides electricity to park m; The heat power demand for integrated energy operator i after heat demand response for community m; is the original heat load power of the community; Reducing heat load for the community;

[0034] Inter-park interaction power constraints:

[0035]

[0036] Among them, P m-n max It represents the maximum limit of the interaction power between community m and community n.

[0037] As a preferred solution of the multi-agent microgrid optimization scheduling method taking into account source-load uncertainty described in the present invention, wherein: the energy management layer analyzes the daily load curve and the power generation curve, extracts key features, and identifies key points of system operation and potential supply-demand imbalances;

[0038] A two-stage multi-time scale rolling optimization model is adopted to consider the timing uncertainty of renewable energy power generation and the uncertainty of power load. In the day-ahead stage, preliminary optimization is carried out to formulate a basic operation plan. In the intraday stage, the optimization results are continuously updated according to real-time data to adapt to the rapid changes in system status.

[0039] Considering the volatility of transaction prices between the main grid and multi-source microgrids, a robust optimization method is used to ensure the economy and stability of the system under price fluctuations. Based on the optimization model and real-time data, the real-time transaction price between the main grid and the multi-source microgrid is calculated and fed back into the target mathematical model of the multi-agent integrated energy system.

[0040] As a preferred solution of the multi-agent microgrid optimization scheduling method taking into account source-load uncertainty described in the present invention, wherein: the two-stage multi-time scale rolling optimization model covers two stages of day-ahead trading and intraday trading, and is optimized at different time scales to adapt to the uncertainty and fluctuation of the system;

[0041] The day-ahead optimization goal is to minimize system operating costs, including electricity and heat costs, equipment operating costs, and wind and solar power abandonment costs;

[0042] Intraday trading adjusts day-ahead plans based on real-time data to manage resources in a refined manner, ensure supply and demand balance, and optimize costs;

[0043] System operation cost function of day-ahead optimal scheduling:

[0044]

[0045] Among them, F m (t), Fp (t) are the equipment operation cost, wind abandonment penalty cost and solar abandonment penalty cost respectively; P e (t), P h (t) are the purchased electricity power and purchased heat power respectively; c e (t), c h (t) represents the unit electricity and heat purchase costs respectively;

[0046] In intraday trading optimization, operators use real-time data to fine-tune day-ahead plans to accommodate forecast deviations, renewable energy fluctuations, and demand changes;

[0047] Intraday optimization minimizes operating costs while reducing scheduling changes, enhancing system robustness, and ensuring supply and demand balance. The objective function is multi-objective, including the same operating costs as the day before, as well as the newly added scheduling adjustment costs, to achieve a balance between cost-effectiveness and system stability. The objective function of the system scheduling adjustment cost is:

[0048]

[0049] Among them, θ1 and θ2 represent the weight factors of the dispatch adjustment cost of the power and thermal systems respectively; P′ load (t), H′ load (t) refer to the power load and thermal load at time t after scheduling optimization;

[0050] In the intraday optimization scheduling stage, a multi-objective optimization strategy is adopted. By assigning specific weights to the objectives of minimizing the operating cost and minimizing the system scheduling adjustment, they are combined into an objective function:

[0051] F=min(λ1F1+λ2F2)

[0052] Among them, λ1 and λ2 are the weights corresponding to the operating cost and the system scheduling adjustment cost respectively; rolling optimization means that in intraday trading, the optimization process will be repeated periodically;

[0053] Each cycle updates the optimization results based on the latest information. The energy management layer transmits the latest scheduling information to the central management layer for further iteration and application to the system.

[0054] The robust optimization model maintains optimality within the fluctuation range of the uncertain variable u; the core goal is to ensure that the strategy is still practically executable even in the most unfavorable situation that may occur in the uncertainty set U. The model is expressed as:

[0055]

[0056] The uncertainty set U and its parameter conditions are expressed as:

[0057]

[0058] Among them, C e,grid (t) and C g,net (t) are the system heat purchase cost and electricity purchase cost from the distribution network during period t; C i,su (t) and C i,sd (t) are the startup and shutdown costs of each generator set in the system during period t; N T The value is 24, which is the time of a day; N is the number of units in the system; P WT (t), P PV (t) and P DR (t) are the actual wind power, solar power output and load response in period t; P f,WT (t), P f,PV (t) and P f,DR (t) are the wind and solar output forecast power and load forecast response in period t respectively; △P WT (t), △P PV (t) and △P DR (t) are wind and solar output prediction errors and load response prediction errors during period t respectively; v + WT 、v - WT 、v + PV 、v - PV 、v + DR 、v - DR is a binary variable. When the value of the corresponding period is 1, the corresponding uncertainty variable is at the boundary of the interval. WT , G PV , G DR are wind power, photovoltaic and load uncertainty adjustment parameters respectively.

[0059] An optimal dispatching system for a multi-agent microgrid taking into account source-load uncertainty using any of the methods described in the present invention, characterized in that:

[0060] The collection unit obtains the parameter information of each energy subject in the energy system, the user's load demand data for energy, and the sensitivity index of each stakeholder;

[0061] A central dispatching unit generates a total objective function according to the sensitivity index and constrains it at the central dispatching layer to construct a decision model of the first layer;

[0062] The aggregation scheduling unit optimizes the output of each energy source through the aggregation scheduling layer during the execution of the overall objective function, calculates the input sensitivity index in operation, and constructs a cooperative game model involving multiple stakeholders in the second layer;

[0063] The energy management unit uses the energy management layer to analyze the wind power, photovoltaic new energy forecast curves and daily load curves, obtains the volatility of the sensitivity indicators when the multi-source microgrid purchases electricity from the main power grid, feeds back the overall objective function, and constructs the third-layer optimization scheduling model; through the three-layer game model, the optimization of the multi-agent integrated energy system is achieved.

[0064] A computer device comprises: a memory and a processor; the memory stores a computer program, wherein: the processor implements the steps of any one of the methods of the present invention when executing the computer program.

[0065] A computer-readable storage medium stores a computer program, wherein: when the computer program is executed by a processor, the steps of any one of the methods of the present invention are implemented.

[0066] Beneficial effects of the present invention: The optimal scheduling method of the multi-agent microgrid taking into account the uncertainty of source-load provided by the present invention can more efficiently utilize renewable energy such as wind energy and solar energy and reduce dependence on traditional fossil energy through the optimization scheduling strategy. The strategy fully considers the intermittent and random nature of new energy power generation, and improves the overall energy utilization efficiency and reduces energy waste through the application of flexible scheduling and energy storage technology. The optimization scheduling strategy improves the system's ability to regulate renewable energy through demand response, energy storage system and other means, reduces the phenomenon of wind and light abandonment, promotes the maximum utilization of renewable energy, and helps to achieve the goal of green and low-carbon energy transformation. On the basis of considering source-load uncertainty and price fluctuations, it provides higher flexibility so that the system can better cope with supply and demand fluctuations. The stability and resilience of the system can be enhanced through a multi-level scheduling optimization method to ensure stable operation under various uncertainties. Through refined scheduling optimization and demand response measures, the strategy can reduce the total operating cost of the energy system under the premise of meeting the system operation constraints. At the same time, through reasonable risk management and price fluctuation response strategies, the impact of price fluctuations on the economy of the system is reduced, and the overall economic benefits are improved. This strategy achieves collaborative optimization of multiple subjects through multi-level optimization of the central dispatching layer, aggregate dispatching layer and energy management layer, improves the interaction efficiency and coordination among various stakeholders, and optimizes the overall performance and benefits of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0068] Figure 1 An overall flow chart of the optimization scheduling method for a multi-agent microgrid taking into account source-load uncertainty provided in the first embodiment of the present invention. DETAILED DESCRIPTION

[0069] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0070] Example 1

[0071] Reference Figure 1 , which is an embodiment of the present invention, provides an optimization scheduling method for a multi-agent microgrid taking into account source-load uncertainty, including:

[0072] S1: Obtain parameter information of each energy subject in the energy system, user load demand data for energy, and sensitivity indicators of each stakeholder.

[0073] The energy entities include three different types of energy entities: wind power, photovoltaic power, and energy storage power station; the parameter information includes power information of charging and discharging of power plants and power stations.

[0074] The sensitivity index of each stakeholder includes a profit index generated when energy circulates between the stakeholder groups, which is used to represent the energy loss after value quantification, and the value of energy when it is acquired is represented by cost, and the value of energy when it is released is represented by price; in the optimization process, the minimum energy loss is achieved by maximizing the profit index.

[0075] S2: At the central scheduling layer, a total objective function is generated according to the sensitivity index and constrained to construct a first-layer decision model.

[0076] The central dispatching layer receives information about wind and solar power abandonment and real-time transaction prices from the energy management layer. Based on the received information and the goal of maximizing the total profit of the system, it optimizes the electricity price structure based on constraints and objective functions to encourage or suppress electricity consumption in a specific time period. By adjusting the electricity price, it guides users to adjust their electricity consumption behavior, such as demand response programs, thereby optimizing the load curve. The optimized electricity price information is passed to the aggregation dispatching layer as an input parameter for its optimized dispatching.

[0077] The interest of MAIES is to maximize the comprehensive profits of the three parks, energy storage power stations and wind farms. Therefore, the mathematical model of MAIES's objectives can be expressed as follows:

[0078]

[0079]

[0080] Among them, F represents the total profit of the multi-agent integrated energy system; N represents the number of 3 parks in this paper; T represents the total scheduling time of 24 hours; C i is the operating profit of park i; C ess is the unit power profit of the energy storage power station; C grid For distribution network profits; is the user's demand response to energy e, where e represents the combination of electricity and thermal loads; is the price at which the park sells energy e to users at time t; is the amount of electricity purchased by park i from the distribution network at time t; represents the electricity selling price of the distribution network at time t; is the power loss of park i at time t; represents the unit profit of the power lost in park i at time t; is the amount of electricity delivered from park i to the energy storage at time t; represents the price of the electricity delivered by park i to the energy storage at time t; The amount of electricity sold by wind power to park i at time t; represents the electricity price of wind power at time t; is the power consumption of device B in park i at time t; is the unit power operation and maintenance cost of device B in park i, covering the operating cost and start-up and shutdown cost; represents the amount of electricity sold by park i to the distribution network at time t; represents the price of electricity sold by park i to the distribution network at time t; The economic compensation provided by the distribution network when responding to the demand of users in park i; is the amount of electricity delivered from the distribution network to the energy storage at time t; Represents the price of electricity delivered by the distribution network to the energy storage at time t; is the amount of electricity sold by wind power to the distribution network at time t.

[0081] The demand response constraint is as follows:

[0082]

[0083] Among them, P DR i,m,t The electric load power after demand response for the park; and are the original electrical load and thermal load power of the community respectively; are the reduced power of electrical load and thermal load in the park; P tran i,m,t The power of the electric load shifted by the community in response to demand response. When the value is positive, it means that the electric load is transferred from other time periods to time period t, and vice versa; r cut is the maximum proportionality factor that can reduce the heat load.

[0084] S3: During the execution of the overall objective function, the output of each energy source is optimized through the aggregation scheduling layer, and the cost-sensitive index in operation is calculated to construct a cooperative game model involving multiple stakeholders in the second layer.

[0085] According to the electricity price information transmitted by the central dispatching layer and the demand-side response measures, the output plans of integrated energy producers and distributors such as wind farms, photovoltaic power stations, and energy storage power stations and energy conversion equipment are optimized according to the dispatching status and dispatching demands to respond to changes in electricity prices and meet load demand. Under the premise of meeting the balance between supply and demand, the production cost is minimized by adjusting the output ratio and time of each energy source. The optimization results of the aggregate dispatching layer are fed back to the central dispatching layer for the next electricity price optimization and evaluation of the total profit of the system.

[0086] Furthermore, a demand response model is established based on the demand.

[0087] Price-based electricity load demand response model:

[0088] By implementing a time-of-use electricity price strategy, residential users who are more sensitive to prices are encouraged to use more electricity during low-price periods and less electricity during high-price periods, thereby smoothing the grid load curve. Formula (5) defines the elasticity of electricity quantity and price. For industrial users, by classifying the electricity load and providing economic compensation based on the amount of load transfer or reduction, they are encouraged to participate in grid load regulation, optimize production plans or use energy storage equipment to achieve load management. Formula (6) is the compensation benefit of incentive-based electricity demand response. This differentiated strategy not only takes into account the sensitivity of residential electricity consumption behavior to electricity price fluctuations, but also takes advantage of the positive response of industrial users to economic incentives, jointly promoting efficient regulation of grid load.

[0089]

[0090] In formula (5), the variable T represents different time points and its value ranges from 1 to n; P n 0 They represent the electricity load at time T and the electricity price at time n before the implementation of the time-of-use electricity price strategy, respectively. T and △P n They represent the load change at time T and the price change at time n after the implementation of demand response intervention. In formula (6), and are the transferable load and curtailable load at time t, and are the compensation unit price per unit electric power transfer and reduction at time t, W s,c To compensate for the benefits of incentive-based electricity demand response; Tn It is the elasticity coefficient, which is used to measure the dependence of user electricity consumption on the increase of electricity price at a certain moment and adjacent moments.

[0091] Heat load demand response model based on fuzzy comfort:

[0092] The heat load demand response model based on fuzzy comfort is a power demand side management strategy that aims to effectively control the heat load in heating by balancing economic incentives and user comfort. The model uses fuzzy logic to deal with the uncertainty of user satisfaction with indoor temperature and converts comfort into a quantitative value. The model considers heat load characteristics, such as thermal inertia and time delay, as well as user sensitivity to temperature changes. By changing electricity prices, users are encouraged to increase heat load during low electricity price periods and reduce usage during peak hours. At the same time, a time series model is used to describe the dynamic changes in the temperature of the heating system. In order to describe the adjustment of the temperature of the heating system over time, Equation (7) uses a time series model to express:

[0093]

[0094] Among them, T h,t , T g,t , T n,t , T w,t They represent the return water temperature, supply water temperature, indoor temperature and outdoor temperature of the heating network at a given time point t, respectively. J refers to the order of the time series model, which is related to the dynamic characteristics of the heating system. ω is the thermal inertia operating parameter of the heating system.

[0095] Energy supply strategy and revenue model of wind farms and photovoltaic farms:

[0096] It is an inevitable direction for the development of the industry to integrate large-scale renewable energy power plants into the centralized and unified dispatch of the power grid as independent profit-making entities. The power generation models of centralized wind farms and centralized photovoltaic farms are shown in formulas (8) and (9), and MAIES is responsible for coordination and control, so as to better integrate them into the centralized power grid dispatch.

[0097]

[0098] P G,t =P e G(t){1-k[T(t)-T e ]} / G e (9)

[0099] In formula (8), P F,t It is expressed as the wind power output at the time, v is the actual measured value of wind speed, and v r , v c Respectively represent the cut-in and cut-out wind speeds, P C is the total installed capacity of wind turbines, v e is the rated wind speed of the system. In formula (9), P G,t is the output power of the photovoltaic device at time t, P e is the maximum output power under rated operating conditions; G(t), G e They correspond to the light intensity at time t and the reference light intensity respectively. k represents the temperature coefficient, which is generally taken as 0.5. T(t) is the photovoltaic surface temperature. T e is the photovoltaic surface temperature under normal conditions. Formula (10) is the profit model of the new energy electric field:

[0100]

[0101] In formula (10), and P i,t w They represent the charging power of the new energy electric field at time t, the power sold by the electric field to the distribution network, and the power of park i respectively; c w,o and W w They are respectively the unit price of electricity sold by new energy power plants, unit power operation and maintenance costs, and profits.

[0102] Energy storage power station charging and discharging strategy and revenue model:

[0103] Energy storage power stations can achieve reasonable use of electricity by charging during periods of low electricity prices (usually at night or off-peak hours) and discharging during periods of high electricity prices (such as daytime or peak hours). The benefits under different operating modes include peak-valley arbitrage, participation in electricity trading, and provision of ancillary services.

[0104] Formula (11) is the profit per unit of energy storage power station.

[0105]

[0106] In the formula, and They represent the charging power of the new energy electric field at time t, the power sold by the electric field to the distribution network, and the power of park i respectively; c w,o and W w They respectively represent the unit price of electricity sold by the new energy power plant, the unit power operation and maintenance cost and profit; n is the value range of time, indicating the total scheduling time of 24 hours; N represents the number of 3 parks in this article.

[0107] The constraints for purchasing electricity and heat are as follows:

[0108]

[0109] Where: P is the power demand of integrated energy operator i after community m completes interaction and demand response with other communities; DR i,m,t The electrical load power after demand response for the community; represents the photovoltaic power generation of the mth park; P m-n i,t P represents the interaction power between the m-park and the n-park served by the energy operator i; m-n i,t >0 means that the m park provides power to the n park, and vice versa. m-n i,t <0 means that park n provides power to park m; The thermal power demand for integrated energy operator i after heat demand response for community m; Q L i,m,t is the original heat load power of the community; Q cut i,m,t Reducing heat load for the community.

[0110] The interactive power constraints between parks are as follows:

[0111]

[0112] Where P m-n max is the maximum limit of the interactive electric power between community m and community n.

[0113] S4: Utilize the energy management layer to analyze the wind power and photovoltaic new energy forecast curves and daily load curves, obtain the volatility of the sensitivity indicators when the multi-source microgrid purchases electricity from the main power grid, provide feedback to the overall objective function, and construct the third-layer optimization scheduling model; through the three-layer game model, the optimization of the multi-agent integrated energy system can be achieved.

[0114] The energy management layer first analyzes the daily load curve and power generation curve to extract key features, such as peak and valley periods, load fluctuations, etc., to identify key points of system operation and potential supply and demand imbalances. A two-stage multi-time scale rolling optimization model is adopted to consider the timing uncertainty of renewable energy power generation and the uncertainty of power load. In the day-ahead stage, preliminary optimization is carried out to formulate a basic operation plan; in the intraday stage, the optimization results are continuously updated according to real-time data to adapt to the rapid changes in system status. In order to cope with the volatility of the transaction price between the main grid and the multi-source microgrid, a robust optimization method is used to ensure the economy and stability of the system under price fluctuations. During the optimization process, the abandoned wind and abandoned light power caused by grid dispatch restrictions or insufficient transaction demand is calculated. According to the optimization model and real-time data, the real-time transaction price between the main grid and the multi-source microgrid is calculated.

[0115] Two-stage multi-time scale rolling optimization model:

[0116] The two-stage multi-timescale rolling optimization model is an effective method for optimizing the dispatch of power and multi-energy systems, covering both day-ahead trading and intraday trading. The model is optimized at different time scales to adapt to the uncertainty and fluctuation of the system.

[0117] The day-ahead optimization goal is to minimize the system operating costs, including electricity and heat costs, equipment operating costs, and wind and solar power abandonment penalties. Intraday trading adjusts the day-ahead plan based on real-time data to manage resources in a refined manner, ensure supply and demand balance, and optimize costs. Therefore, the system operating cost function of the day-ahead optimal scheduling is shown in formula (14):

[0118]

[0119] Where: F m (t), F p (t) are the equipment operation cost, wind abandonment penalty cost and solar abandonment penalty cost respectively; P e (t), P h (t) are the purchased electricity power and purchased heat power respectively; c e (t), c h (t) represent the unit cost of electricity and heat purchase respectively.

[0120] In intraday trading optimization, operators use real-time data to fine-tune the day-ahead plan, usually in units of hours or half hours, to adapt to forecast deviations, renewable energy fluctuations, and demand changes. Intraday optimization minimizes operating costs while reducing scheduling changes, enhancing system robustness, and ensuring supply and demand balance. The objective function is multi-objective, including the same operating costs as the day-ahead, as well as the newly added scheduling adjustment costs, aiming to achieve a balance between cost-effectiveness and system stability. The objective function of the system scheduling adjustment cost is shown in formula (15):

[0121]

[0122] Where: θ1 and θ2 represent the weight factors of the dispatch adjustment cost of the power and thermal systems respectively; P′ load (t), H′ load (t) refer to the power load and thermal load at time t after scheduling optimization.

[0123] In the intraday optimization scheduling stage, a multi-objective optimization strategy is adopted. This paper combines the objectives of minimizing operating costs and minimizing system scheduling adjustments into a single optimization problem by assigning specific weights to them. The combined objective function expression is as follows:

[0124] F=min(λ1F1+λ2F2) (16)

[0125] Where: λ1 and λ2 are the weights corresponding to the operating cost and the system scheduling adjustment cost respectively. Rolling optimization means that the optimization process is repeated periodically in intraday trading. Each cycle will update the optimization results based on the latest information. The energy management layer will transmit the latest scheduling information to the central management layer for further iteration and apply it to the system. Through the two-stage rolling optimization scheduling model of day-ahead and intraday, the peak-shaving resources of the system can be reasonably scheduled, the peak-valley difference of the system can be reduced, and the economy and reliability of the system operation can be ensured.

[0126] Furthermore, the robust optimization model strategy needs to maintain its optimality within the fluctuation range of the uncertain variable u. The core goal is to ensure that the strategy is still practical even in the most unfavorable situation that may occur in the uncertainty set U. The model is as follows:

[0127]

[0128] The uncertainty set U is shown in formula (18), and the parameter conditions are shown in formula (19):

[0129]

[0130] Among them, C e,grid (t) and C g,net(t) are the system heat purchase cost and electricity purchase cost from the distribution network during period t; C i,su (t) and C i,sd (t) are the startup and shutdown costs of each generator set in the system during period t; N T The value is 24, which is the time of a day; N is the number of units in the system; P WT (t), P PV (t) and P DR (t) are the actual wind power, solar power output and load response in period t; P f,WT (t), P f,PV (t) and P f,DR (t) are the wind and solar output forecast power and load forecast response in period t respectively; △P WT (t), △P PV (t) and △P DR (t) are wind and solar output prediction errors and load response prediction errors during period t respectively; v + WT 、v - WT 、v + PV 、v - PV 、v + DR 、v - DR is a binary variable. When the value of the corresponding period is 1, the corresponding uncertainty variable is at the boundary of the interval. WT , G PV , G DR are wind power, photovoltaic and load uncertainty adjustment parameters respectively.

[0131] On the other hand, this embodiment also provides an optimization dispatching system for a multi-agent microgrid taking into account source-load uncertainty, which includes:

[0132] The collection unit obtains the parameter information of each energy subject in the energy system, the user's load demand data for energy, and the sensitivity index of each stakeholder.

[0133] The central dispatching unit generates a total objective function according to the sensitivity index at the central dispatching layer and constrains it, thereby constructing a decision model of the first layer.

[0134] The aggregation scheduling unit, during the execution of the overall objective function, optimizes the output of each energy source through the aggregation scheduling layer, calculates the cost-sensitive index in operation, and constructs a cooperative game model involving multiple stakeholders at the second layer.

[0135] The energy management unit uses the energy management layer to analyze the wind power, photovoltaic new energy forecast curves and daily load curves, obtains the volatility of the sensitivity indicators when the multi-source microgrid purchases electricity from the main power grid, feeds back the overall objective function, and constructs the third-layer optimization scheduling model; through the three-layer game model, the optimization of the multi-agent integrated energy system is achieved.

[0136] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0137] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0138] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0139] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0140] Example 2

[0141] The following is an embodiment of the present invention, which provides an optimization scheduling method for a multi-agent microgrid taking into account source-load uncertainty. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0142] By comparing the energy loss before and after using the present invention, the beneficial effect of the present invention is verified. Table 1 is obtained by testing the same system under the same environment.

[0143] Table 1 Data comparison table

[0144]

[0145]

[0146] In the table, the loss coefficient represents the value loss of the energy of the equipment itself. It can be seen that the present invention significantly reduces energy loss, improves system efficiency and the conversion efficiency of various equipment, optimizes the operation of the comprehensive energy system, and has significant technical advantages and application value through the three-layer optimization scheduling strategy.

[0147] 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 preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. Optimal dispatching method for multi-agent microgrid considering source-load uncertainty, It is characterized in that include: Obtain parameter information of each energy subject in the energy system, user load demand data for energy, and sensitivity indicators of each stakeholder; At the central dispatching layer, a total objective function is generated according to the sensitivity index and constrained to construct a first-layer decision model; In the process of executing the total objective function, the output of each energy source is optimized through the aggregation scheduling layer, and the cost-sensitive index in operation is calculated to construct a cooperative game model involving multiple stakeholders in the second layer; The energy management layer is used to analyze the wind power, photovoltaic new energy forecast curves and daily load curves to obtain the volatility of the sensitivity indicators when the multi-source microgrid purchases electricity from the main power grid. The overall objective function is fed back to construct the third-layer optimization scheduling model; through the three-layer game model, the optimization of the multi-agent integrated energy system is achieved.

2. The optimization scheduling method of a multi-agent microgrid taking into account source-load uncertainty as claimed in claim 1, Features: The energy entities include three different types of energy entities: wind power, photovoltaic power, and energy storage power station; the parameter information includes power information of charging and discharging of power plants and power stations; The sensitivity index of each stakeholder includes a profit index generated when energy circulates between the stakeholder groups, which is used to represent the energy loss after value quantification, and the value of energy when it is obtained is represented by cost, and the value of energy when it is released is represented by price; in the optimization process, the minimum energy loss is achieved by maximizing the profit index; The central dispatching layer receives information on abandoned wind and solar power and real-time transaction prices from the energy management layer, and based on the received information, constructs constraints and objective functions to optimize the electricity price structure with the goal of maximizing the total profit of the system.

3. The optimization scheduling method of a multi-agent microgrid taking into account source-load uncertainty as claimed in claim 2, Features: The objective function includes: the interest demands of the multi-agent integrated energy system are expressed as the maximization of the comprehensive profits of the three parks, energy storage power stations and wind farms; The objective mathematical model of the multi-agent integrated energy system is expressed as: Among them, F represents the total profit of the multi-agent integrated energy system; N represents the number of three parks in this paper; T represents the total scheduling time of 24 hours; C i is the operating profit of park i; C ess is the unit power profit of the energy storage power station; C grid is the profit of the distribution network; i,t load,DR is the user's demand response to energy e, where e represents the combination of electricity and thermal loads; c t e,sel P is the price at which the park sells energy e to users at time t; i,t grid,buy is the amount of electricity purchased from the distribution network by park i at time t; c t grid,sel represents the electricity price of the distribution network at time t; P i,t dis is the power loss of park i at time t; c t ess,dis represents the unit profit of the power loss of park i at time t; P i,t ch is the amount of electricity delivered from park i to the energy storage at time t; c t ess,ch represents the price of electricity delivered by park i to the energy storage at time t; P i,t wind c is the amount of electricity sold by wind power to park i at time t; t wind,sel represents the electricity price of wind power at time t; P i,t B is the power consumption of device B in park i at time t; C i B,op is the unit power operation and maintenance cost of device B in park i, including operation cost and start-up and shutdown cost; P i,t grid,sel is the amount of electricity sold by park i to the distribution network at time t; c t ex,sel represents the price of electricity sold by park i to the distribution network at time t; C i e,DR The economic compensation provided by the distribution network when responding to the needs of users in park i; P t ch ,grid is the amount of electricity delivered from the distribution network to the energy storage at time t; c t ess,grid represents the price of electricity delivered by the distribution network to the energy storage at time t; P t wind The amount of electricity sold by wind power to the distribution network at time t; The demand response constraint is expressed as: Among them, P DR i,m,t P is the electric load power after the park responds to demand; L i,m,t and Q L i,m,t are the original electric load and heat load power of the community respectively; P cut i,m,t , Q cut i,m, t are the reduced power of the park’s electrical load and thermal load respectively; P tran i,m,t The power of the electric load shifted by the community in response to demand response. When the value is positive, it means that the electric load is transferred from other time periods to time period t, and vice versa; r cut is the maximum proportionality factor that can reduce the heat load.

4. The optimization scheduling method of a multi-agent microgrid taking into account source-load uncertainty as claimed in claim 3, Features: The aggregate dispatching layer obtains the electricity price information and demand-side response measures transmitted by the central dispatching layer; optimizes the output plans of wind farms, photovoltaic power stations, energy storage power stations, integrated energy producers and distributors, and energy conversion equipment according to the dispatching status and dispatching demands, so as to respond to changes in electricity prices and meet load demands; On the premise of meeting the supply and demand balance, the production cost is minimized by adjusting the output ratio and time of each energy source. The optimization results of the aggregation scheduling layer are fed back to the central scheduling layer for the next electricity price optimization and evaluation of the total system profit.

5. The optimization scheduling method of a multi-agent microgrid taking into account source-load uncertainty as claimed in claim 4, Features: In the above-mentioned aggregation scheduling layer, the revenue model of the new energy power plant is: Among them, P t c,w , P t w,w and P i,t w They represent the charging power of the new energy electric field at time t, the power sold by the electric field to the distribution network, and the power sold by the park i; c t w,s 、c w,o and W w They represent the unit price of electricity sold by new energy power plants, the unit power operation and maintenance cost and profit respectively; n is the value range of time, indicating the total scheduling time of 24 hours; N represents the number of 3 parks in this article; Energy storage power station unit profit model: Among them, P i,t F , P i,t C Respectively represent the discharge and charging power of the energy storage power station to park i; P t w,w and P t c,w are the charging power obtained by the energy storage power station from the upper power grid and the centralized wind farm; c t e represents the comprehensive price of electricity sold by the energy storage power station; ψ and W z,s Respectively represent the unit power operation and maintenance cost and profit of the energy storage power station; Constraints on purchasing electricity and heat: Among them, P load i,m,t P is the power demand of integrated energy operator i after community m completes interaction and demand response with other communities; DR i,m,t The electric load power after the community performs demand response; P PV i,m,t represents the photovoltaic power generation of the mth park; if P m-n i,t >0, it means that the m park provides power to the n park. If P m-n i,t <0, it means that park n provides power to park m; Q load i,m,t The thermal power demand for integrated energy operator i after heat demand response for community m; Q L i,m,t is the original heat load power of the community; Q cut i,m,t Reducing heat load for the community; Inter-park interaction power constraints: Among them, P m-n max It represents the maximum limit of the interaction power between community m and community n.

6. The optimization scheduling method of a multi-agent microgrid taking into account source-load uncertainty as claimed in claim 5, Features: The energy management layer analyzes the daily load curve and power generation curve, extracts key features, and identifies key points of system operation and potential supply and demand imbalances; A two-stage multi-time scale rolling optimization model is adopted to consider the timing uncertainty of renewable energy power generation and the uncertainty of power load. In the day-ahead stage, preliminary optimization is carried out to formulate a basic operation plan. During the intraday stage, the optimization results are continuously updated based on real-time data to adapt to the rapid changes in system status; Considering the volatility of transaction prices between the main grid and multi-source microgrids, a robust optimization method is used to ensure the economy and stability of the system under price fluctuations; According to the optimization model and real-time data, the real-time transaction price between the main grid and the multi-source microgrid is calculated and fed back into the target mathematical model of the multi-agent integrated energy system.

7. The optimization scheduling method of a multi-agent microgrid taking into account source-load uncertainty as claimed in claim 6, Features: The two-stage multi-timescale rolling optimization model covers two stages, day-ahead trading and intraday trading, and is optimized at different time scales to adapt to the uncertainty and volatility of the system; The day-ahead optimization goal is to minimize system operating costs, including electricity and heat costs, equipment operating costs, and wind and solar power abandonment costs; Intraday trading adjusts day-ahead plans based on real-time data to manage resources in a refined manner, ensure supply and demand balance, and optimize costs; System operation cost function of day-ahead optimal scheduling: Among them, F m (t), F p (t) are the equipment operation cost, wind abandonment penalty cost and solar abandonment penalty cost respectively; P e (t), P h (t) are the purchased electricity power and purchased heat power respectively; c e (t), c h (t) represents the unit electricity and heat purchase costs respectively; In intraday trading optimization, operators use real-time data to fine-tune day-ahead plans to accommodate forecast deviations, renewable energy fluctuations, and demand changes; Intraday optimization minimizes operating costs while reducing scheduling changes, enhancing system robustness, and ensuring supply and demand balance. The objective function is multi-objective, including the same operating costs as the day before, as well as the newly added scheduling adjustment costs, to achieve a balance between cost-effectiveness and system stability. The objective function of the system scheduling adjustment cost is: Among them, θ 1 ,θ 2 Respectively represent the weight factors of the dispatch adjustment cost of the power and thermal systems; P′ load (t), H′ load (t) refer to the power load and thermal load at time t after scheduling optimization; In the intraday optimization scheduling stage, a multi-objective optimization strategy is adopted, by assigning specific weights to the objectives of minimizing the operating cost and minimizing the system scheduling adjustment, and merging them into the objective function: F=min(λ 1 F 1 +λ 2 F 2 ) Among them, λ 1 , 2 are the weights corresponding to the operating expenses and the system scheduling adjustment costs respectively; rolling optimization means that the optimization process will be repeated periodically in intraday trading; Each cycle updates the optimization results based on the latest information. The energy management layer transmits the latest scheduling information to the central management layer for further iteration and application to the system. The robust optimization model maintains optimality within the fluctuation range of the uncertain variable u; the core goal is to ensure that the strategy is still practically executable even in the most unfavorable situation that may occur in the uncertainty set U. The model is expressed as: The uncertainty set U and its parameter conditions are expressed as: Among them, C e,grid (t) and C g,net (t) are the system heat purchase cost and electricity purchase cost from the distribution network during period t; C i,su (t) and C i,sd (t) are the startup and shutdown costs of each generator set in the system during period t; N T The value is 24, which is the time of a day; N is the number of units in the system; P WT (t), P PV (t) and P DR (t) are the actual wind power, solar power output and load response in period t; P f,WT (t), P f,PV (t) and P f,DR (t) are the wind and solar output forecast power and load forecast response in period t respectively; △P WT (t), △P PV (t) and △P DR (t) are respectively the wind and solar output prediction errors and load response prediction errors in period t; is a binary variable. When the value of the corresponding period is 1, the corresponding uncertainty variable is at the boundary of the interval. WT , G PV , G DR are wind power, photovoltaic and load uncertainty adjustment parameters respectively.

8. An optimization dispatching system for a multi-agent microgrid taking into account source-load uncertainty using the method as claimed in any one of claims 1 to 7, Features: The collection unit obtains the parameter information of each energy subject in the energy system, the user's load demand data for energy, and the sensitivity index of each stakeholder; A central dispatching unit generates a total objective function according to the sensitivity index and constrains it at the central dispatching layer to construct a decision model of the first layer; The aggregation scheduling unit optimizes the output of each energy source through the aggregation scheduling layer during the execution of the overall objective function, calculates the input sensitivity index in operation, and constructs a cooperative game model involving multiple stakeholders in the second layer; The energy management unit uses the energy management layer to analyze the wind power, photovoltaic new energy forecast curves and daily load curves, obtains the volatility of the sensitivity indicators when the multi-source microgrid purchases electricity from the main power grid, feeds back the overall objective function, and constructs the third-layer optimization scheduling model; through the three-layer game model, the optimization of the multi-agent integrated energy system is achieved.

9. A computer device, include: Memory and processor; The memory stores a computer program, characterized in that: when the processor executes the computer program, the processor implements the steps of the optimization scheduling method of the multi-agent microgrid taking into account source-load uncertainty.

10. A computer-readable storage medium having a computer program stored thereon, Features: When the computer program is executed by a processor, the steps of the method for optimizing scheduling of a multi-agent microgrid taking into account source-load uncertainty are implemented.

Citation Information

Cited By

  • High-energy-consumption industrial micro-grid hierarchical scheduling optimization method based on load classification

    CN119647844A

  • Airport energy balance management method, system, equipment and medium

    CN120235415A

  • Intelligent substation energy management method based on cloud computing

    CN120657861A

  • A cloud computing-based intelligent substation energy management method

    CN120657861B

  • Regional optical storage and charging micro-grid control method and system

    CN120749789A