Multi-microgrid bid invitation energy management method based on energy router

Abstract

The invention relates to the technical field of micro-grids, in particular to a multi-micro-grid bid invitation energy management method based on an energy router, which comprises the following steps of: firstly, providing a multi-micro-grid structure comprising the energy router, and regarding multiple MGER as a multi-agent system of an intelligent individual consisting of the energy router; an energy router is considered to be a public information and energy flow platform that shares its activity information with a power distribution network center. Energy conservation and efficiency improvement are realized by utilizing the characteristics of bidirectional energy flow, information monitoring and remote operation, the flexibility and flexibility are increased, the influence of uncertain factors such as wind driven generator and photovoltaic output in the microgrid on energy supply is considered, a reasonable collaborative optimization operation strategy is realized, and the energy supply efficiency is improved. Therefore, the energy router with ordered and reasonable energy interaction is ensured to maximize social profit and ensure fair income distribution under the background of uncertain wind energy and solar energy output.

Description

Technical Field

[0001] The present invention relates to the field of microgrid technology, and in particular to a multiple microgrid bidding energy management method based on energy routers. Background Art

[0002] With the vigorous development of society, the level of industrialization of mankind is constantly improving, and the problem of environmental pollution is becoming more and more serious. In order to solve the contradiction between human energy utilization and living environment, governments around the world are vigorously promoting the development of renewable energy. In order to promote clean and low-carbon energy transformation, advance energy revolution, and improve energy efficiency, it is urgent to transform the traditional power system into a new smart power system to meet the growing demand for clean and renewable energy. The new smart power system can effectively integrate renewable energy, thereby promoting the development of clean energy. In the new smart power system, actively developing smart microgrids that mainly consume new energy to achieve compatibility and complementarity with large power grids has become the focus of promoting the development of new power systems. In the new smart microgrid, the power market is no longer monopolized by a few large power companies, but uses advanced power grids. Electronics and information and communication technology (ICT) realize distributed energy trading. Therefore, the emerging concept of Energy Internet (IoE) is proposed. Through digital, intelligent and integrated technical means, efficient scheduling, management and trading of energy can be effectively realized. Energy router (ER) as the execution terminal of IoE can provide plug-and-play ports for distributed power sources, energy storage, loads, etc., and provide continuous power supply for important loads in the area when the distribution network is out of power. At the same time, ER can serve as the core control device of energy microgrid to control the flow to adapt to the ever-changing energy demand and achieve scale. Therefore, studying the collaboration between microgrid and energy router (MGER) in IoE is of great significance to promoting innovation and progress in the energy industry, improving the security and flexibility of energy supply and building a clean, efficient and sustainable energy system.

[0003] A decentralized Dijkstra algorithm is currently proposed to solve the energy routing problem between multiple energy trading pairs. Each designated energy trading pair uses the Dijkstra algorithm to calculate the optimal transmission line from its own perspective. However, it can only generate a solution for a single path. In order to reduce loss, some transmission lines with smaller capacity may be ignored, which may cause network congestion. The operation of the energy Internet is explored, and a routing algorithm for finding the shortest loss path for energy routers is introduced. First, the source router and the destination router are selected, and the algorithm is used to identify different paths between the two routers to reduce network congestion and enable the energy router microgrid to reasonably distribute energy.

[0004] The operation mechanism of the above-mentioned integrated energy microgrid group is relatively complex, and does not take into account the organic connection between MGER and the upper-level microgrid, which makes the collaborative optimization operation between microgrid groups difficult and is not conducive to the development of an integrated energy microgrid group system containing energy routers.

[0005] In view of this, the present invention proposes a multi-microgrid bidding energy management method based on energy router to solve the above problems. Summary of the Invention

[0006] In view of the shortcomings of the existing technology, the purpose of the present invention is to propose a multi-microgrid bidding energy management method based on energy routers, which has the advantages of reducing network congestion and enabling energy router microgrids to reasonably distribute energy, and solves the problem that the energy router lacks flexibility and scalability.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A multi-microgrid bidding energy management method based on energy routers includes the following steps:

[0009] Firstly, a multi-microgrid structure including energy routers is proposed. It regards multiple MGERs as a multi-agent system composed of intelligent individuals composed of energy routers. The energy router is considered as a public information and energy flow platform that shares its activity information with the distribution network center. Then, based on this, an integrated energy microgrid optimization operation strategy based on conditional value at risk (CVaR) and Nash negotiation model is proposed.

[0010] A CVaR-based master-slave profit allocation model is then established to measure the risk of wind and solar uncertainty. The contribution of microgrids to social welfare is measured by the degree of MGER transaction cost loss. Each microgrid conducts asymmetric negotiations based on the scale of its social profit contribution to achieve a fair distribution of cooperative surplus.

[0011] The effectiveness of the proposed method is verified through case analysis. Simulation results show that the proposed method can maximize the social profit of MGER in a multi-microgrid structure. Based on the cooperative benefits of the microgrid alliance, MGER can achieve fair energy distribution in terms of social profit contribution. This provides a feasible reference scheme for the operation decision-making of energy router multi-microgrid systems between economic benefits and risk levels.

[0012] The interaction pattern between the energy microgrid and the distribution network center is represented as a Stackelberg game model. As a leader, the distribution network center provides pricing decisions, enabling the energy microgrid to participate in energy management and achieve maximum social profit. As an energy router that tracks the changes in the distribution network center's decisions, it has only one "black box" of output and use. Through cooperation in the distribution network center, energy configuration is shared. As a leader, the distribution network center needs to make a choice between expected profit and CVaR to maximize social profit. Multiple microgrid structures with energy routers;

[0013] The distribution network center will incorporate the energy routers of all participants into the "black box" of the "power group" for unified management to achieve optimal energy supply between the interconnected multi-energy systems (MES). In addition, the uncertainty brought by potential risks, energy routers, microgrids, wind power and solar power generation, in order to improve the practicality of the model, and the uncertainty brought by risk factors are taken into account in the distribution network center. The interconnected multi-energy system (MES) can control the overall risk level from a holistic perspective;

[0014] As a leader, the distribution network coordinates the supply and demand of the microgrid. When the microgrid has energy procurement needs, the distribution network needs to provide energy. When the microgrid generates too much energy, the distribution network needs to purchase electricity to increase the penetration rate of new energy in the microgrid itself. When there is a mismatch within the microgrid, it requires retailers to coordinate with the distribution network. Generally speaking, the sales price is lower than the purchase price to encourage production and consumers to arrange internal energy. However, during the coordination process, the microgrid itself cannot obtain energy unlimitedly during low-price periods, and the distribution network itself is not an infinitely functional system, which may lead to the collapse of the power grid. Therefore, in order to achieve coordinated management and reduce the impact on the distribution network (DN), the goal of retailers is to control the risk level and cope with uncertainty while maximizing social profits.

[0015] As an energy router microgrid that changes with the response of the distribution network, the energy router participates in the production and distribution of energy. Therefore, the energy router is called a producer and seller. As a follower that changes with the response of the distribution network, the energy router determines its own transactions based on changes in risks and operating costs;

[0016] MES is jointly established by multiple microgrids containing ER. When configuring, the distribution center only needs to consider the total configuration cost of MES and the risk operation cost brought by CVAR calculation. The distribution center needs to reasonably configure MES to minimize the operating cost of the system. At the same time, each microgrid containing ER will adjust the output of various devices in the system according to the configuration of the distribution center to flatten the net load. Therefore, this problem is a two-level optimization problem. This paper requires the use of a solver or heuristic algorithm to solve the problem of maximizing the overall social profit of multiple microgrids. The microgrid model and the distribution network model are non-convex linear models. In order to facilitate the solution, in the non-convex problem of joint optimization of multiple energy microgrids, these nonlinear constraints are relaxed and the problem is decomposed into two sub-problems to solve, namely solving the global cost minimization sub-problem (P1) and the energy greed acquisition sub-problem (P2) tracking the distribution company to complete market settlement, so that the problem can be solved by the solver;

[0017] By relaxing the nonlinear terms, problems in local optimality can be transformed into global optimality. Additional augmented variables need to be introduced for expansion. Then, the spatial branch and bound method can be used. This is the most general algorithm for solving mixed integer nonlinear programming (MINLP). It is different from the traditional branch and bound method for solving mixed integer linear programming (MILP) problems. Relaxation is no longer a simple linear relaxation, but an approximate function that uses a convex underestimater or concave overestimater for each constraint and optimization objective, providing lower / double bound pruning. The branch and bound method not only needs to reduce the upper and lower bounds of the integer variables, but also needs to restrict the variables included in the constraints and update the corresponding approximate functions. In the process of relaxing variables, the constraints on energy trading are mainly discussed. In the energy trading process, due to the mutual exclusion between the same buyer and seller, they cannot occur simultaneously;

[0018] The complementary relaxation condition of the lower energy level model can be expressed as:

[0019]

[0020]

[0021]

[0022]

[0023]

[0024]

[0025]

[0026]

[0027]

[0028]

[0029] At the optimal solution point of the lower model, we have

[0030]

[0031]

[0032]

[0033]

[0034] The original two-level collaborative planning model can be transformed into a single-level MPEC problem, making it a non-convex model that is not suitable for direct solution. In this section, we will use the Big-M method to apply convex constraints to the non-convex part of the above model and linearize it using strong duality theory;

[0035] Using the Big-M method, by introducing several 0-1 variables, the original nonlinear constraint is equivalently converted into a mixed integer linear constraint. This paper discusses the constraint problem of ET. In the ET process, it is impossible to sell and buy energy for the same buyer and seller at the same time. Therefore, no matter whether the energy is increasing or decreasing, there exists a transmission and distribution model as shown in the formula;

[0036]

[0037]

[0038]

[0039]

[0040] It is the energy that producers and sellers sell to the distribution network; is the energy purchased by producers and sellers from the distribution network; M is a sufficiently large constant;

[0041] Based on the strong duality theory, the bilinear terms can be linearized, and the previous model can be transformed into a mixed integer second-order cone programming (MISOCP) model that can be optimized and solved. The objective function is shown as follows:

[0042]

[0043] Commercial solvers are available for this convex optimization problem and have been market approved;

[0044] Perform scenario generation and restoration, followed by CVAR calculation. Cornish-Fisher is a continuous asymptotic expansion method for the quantile of a standardized random variable. It expands the quantile function of a real distribution into a polynomial function of the quantile function of a standard normal distribution. The first four orders of expansion are expressed as follows:

[0045]

[0046] Where z α is the quantile α of the standard normal distribution, κ3 and κ4 represent the skewness and kurtosis respectively. Under this quantile expansion, the loss random variable can be expressed as:

[0047]

[0048] Where μ l and σ l They represent the expected value and standard deviation of the loss random variable in period l, respectively. It can be seen that without knowing the true distribution of the loss random variable, the skewness κ3 and kurtosis κ4 are estimated from the sample, and the asymptotic expansion and value of the quantile are obtained;

[0049] Considering the potential uncertainties in multiple aspects of energy supply, consumption, and transaction, which increase the risk of distribution network operation, I represents the number of microgrids. The risk of multiple uncertainties in the real-time stage can be expressed as Equation (44):

[0050]

[0051] Among them, s1, s2, ..., sN are sample data, α is the confidence level, and η i,w ,η i,v Calculate CVaR using auxiliary variables related to consumer i and scenario w;

[0052] In order to avoid false regression and eliminate heteroscedasticity, it is usually necessary to take the logarithmic wind and solar series time attributes of the data without changing the attributes and correlation of the time series. In order to obtain smooth data and make the data closer to normal distribution, the formula for taking the series is:

[0053] rt=-log(p)+log(p)

[0054] It can be seen that most r are smaller than the corresponding VR and CVaR estimates, that is, the risk estimates are essentially valid. If r is greater than the corresponding VaR or CVaR value, the risk estimation fails. Obviously, CVaR is more reliable for the data. Then, compared with the theoretical value and the CVaR method, the VaR method is obviously not suitable for risk measurement in the power bidding market. The CVaR method estimates market risk well, while the VaR method has significant differences from the actual situation when estimating risk.

[0055] Case Analysis

[0056] This article uses Figure 1 The three-node model in is given, and the parameters of the three microgrids are given for arithmetic calculations. The parameters of these devices are set according to Peng et al. (2021) and Na et al. (2023);

[0057] CVaR Impact Analysis

[0058] Calculate the CVaR and expected profit at 95% confidence level for different risk factors by modifying the parameters in the equation, β;

[0059] The expected profits and CVaR of producers and sellers with different β values ​​should be listed as: (a) expected profits and CVaR1 of producers and sellers; (b) expected profits and CVaR2 of producers and sellers; (c) expected profits and CVaR3 of the market. The first point is that by solving the parameter β close to 0, which represents the maximum profit, producers and sellers may be risk-averse and approximate the optimal trading strategy of producers and sellers when they do not consider CVaR. It can be seen that as the bidding degree increases, producers and sellers become more and more risk-averse, and the expected returns decrease accordingly. The distribution network, as a leader, takes the maximization of overall profits as the overall goal, and analyzes the impact of different risk indicators on the trading profits of each producer and seller, thereby guiding the output of producers and sellers of new energy transactions and promoting the amount of new energy consumption;

[0060] Simulation results analysis

[0061] The transaction price between the distribution network and buyers and sellers depends on their attitude towards risk, which is caused by the uncertainty of wind and solar energy output. Retailers determine reasonable prices within a price range by setting minimum and maximum values ​​to maximize social welfare. Therefore, risk-seeking retailers may offer competitive prices to attract prosumers to participate in energy sharing while taking on relatively high risks.

[0062] Iterative Analysis

[0063] The model is solved using the CPlex solver. The optimal risk transaction value of each producer and seller is obtained through sorting and looping algorithms, and the optimal price curve that maximizes overall welfare is obtained. The method converges after 86 iterations, with a computation time of 689 seconds. After meeting the iterative convergence result, the residual error of each microgrid iteration converges to within 10-3. Therefore, the method proposed in this paper has good convergence performance and computational efficiency, meeting the needs of optimized scheduling.

[0064] Negative prices represent the energy selling prices of producers and sellers, while positive prices represent the purchase prices of energy. For profit, when distribution network operators sell energy to producers and sellers, the sales price is often higher than the purchase price. However, due to the load price limit, the purchase price cannot be higher than the load price, making it profitable for producers and sellers. Positive prices represent the purchase quantity of producers and sellers, while negative prices represent the sales quantity.

[0065] Draw conclusions

[0066] This paper proposes a two-layer energy interaction framework for multiple microgrids with energy routers, which can help the distribution network center coordinate energy transactions with energy routers. The multi-microgrid containing energy routers is regarded as a multi-agent system, and its activity information can be shared with the distribution network center. On this basis, we propose a static game theory model with the distribution network center as the leader and the microgrid containing energy routers as the follower. At the same time, risk measurement methods - VaR and CVaR calculations - are introduced to identify and quantify potential losses under the uncertainty of wind and solar output. The leader makes a balanced choice between income and risk, determines price conditions under the current risk indicators, encourages followers to engage in energy transactions, expands wind and solar energy consumption, and improves the overall return of the microgrid group. Considering the wind and solar power generation proposed in this paper In order to deal with the uncertainty of electricity, in the multi-microgrid bidding strategy using energy routers, energy management is divided into two stages. In the first stage, a master-slave profit distribution model based on the conditional risk value is established based on the interest needs of each microgrid and the energy supply and demand relationship, realizing the risk measurement of the dual uncertainty of wind and solar energy. In the second stage, the above problem is decomposed into the minimum cost sub-problem and the maximum transaction quantum problem of asymmetric negotiation, and energy is fairly distributed according to the contribution value of each microgrid containing energy routers. The simulation results show that the model enhances the ability of multiple microgrid systems with energy routers to cope with uncertain wind and solar energy output, improves the effectiveness and rationality of energy interaction between microgrid energy routers and supporting networks, and maximizes the profits of multiple microgrids, ensuring fair distribution of stakeholders.

[0067] Future work will include analyzing and formulating optimization problems for microgrids with energy routers in different configurations, further considering classification studies of energy routers in microgrids, and determining the impact of different microgrids on multiple microgrid systems to improve the widespread and universal application of energy routers in multiple microgrid systems.

[0068] As a preferred technical solution, this strategy aims to achieve the effectiveness and rationality of the energy interaction between MGER and the distribution network, while maximizing social welfare and ensuring fairness in income distribution.

[0069] As a preferred technical solution, in order to clearly describe the coupling relationship of various energy sources in energy routers under multiple microgrids, a matrix method is used from the perspective of port devices to describe the mutual relationship and interaction pattern of electric energy in energy routers. By connecting the input, output and conversion of electric energy together and imposing constraints to simplify the calculation of the distribution network center, it can simultaneously meet the energy supply and distribution needs of microgrids in different periods.

[0070] As a preferred technical solution, producers and sellers can cooperate with other producers and sellers to reduce operating costs. At the same time, producers and sellers maximize the use of energy they can obtain from the distribution network under limited constraints. This can be called greed. Therefore, we decompose the optimal scheduling problem into the minimum cost subproblem P1 and the maximum transaction quantum problem P2. The calculation formula is

[0071] As the preferred technical solution, the maximum acquisition transaction volume is: When the distribution network determines supply and price, the market electricity price will be affected, which will in turn affect the decision-making of electricity traders and consumers, and make appropriate negotiation decisions through formulas. The decision of the distribution network will affect the decision of the microgrid, and then affect the decision of the distribution network layer. This feedback loop can lead to adjustments in market prices and supply and demand relationships, and ultimately tend to the optimal equilibrium state.

[0072] As a preferred technical solution, the energy conversion matrix model is shown as the equation Z=XI Where Z, X, and I represent the port output power matrix, conversion coupling matrix, and port input power matrix of the energy router. In the conversion coupling matrix X, the diagonal X nn Indicates the network loss coefficient and energy distribution coefficient in the conversion, the diagonal X mn It represents the energy conversion coefficient s between device m and device n. The conversion coefficient includes the energy distribution coefficient and the energy efficiency coefficient. a represents the energy distribution coefficient. If the sum of the coefficients of the devices at the same port is 1, b is the port energy conversion efficiency, and c is the energy transmission loss coefficient.

[0073] As the preferred technical solution, the decision model of the distribution network is as follows: Where i represents the measure of producers and sellers; t represents the measure of time; N represents the total number of producers and sellers; T represents the sum of all time; Ω represents the total capacity of producers and sellers; π i represents the probability that producers and sellers are willing to take risks when purchasing electricity, β is the weighted coefficient of the profit given up by producers and sellers when load is reduced, and ζ i Auxiliary calculation of the value at risk (VaR) associated with the producer-seller, γ and the constraints associated with the Lagrange multiplier operator; η i,w ,η i,v Wind and sunlight condition values ​​and risk indicators, the price of energy purchased by producers and sellers from the distribution grid; the price at which producers and distributors sell energy to the distribution network, the price at which producers and distributors sell energy to the distribution network, Producers and sellers sell energy to the distribution grid, Producers and sellers purchase energy from the distribution network. The upper and lower limits on energy sales by producers and sellers in the distribution network, Floors and floors for purchasing energy from producers and sellers in the distribution network New York, the average price at which energy is sold by producers and distributors to consumers; the average price at which producers and distributors sell energy to the distribution network;

[0074] Expected retailer revenue (expected sales revenue to production consumers minus the expected cost of purchasing from DN), the trade-off between expected revenue and risk (CVaR multiplied by the weighting parameter β), the larger the β value, the greater the preference for risk aversion, and a lower β value is chosen to obtain higher revenue to reduce expected costs. In contrast, conservative retailers tend to choose a larger β value to increase the weight of risk aversion, and limit the transaction price offered to consumers to a time interval. In order to limit the market power of retailers, a daily average price is set to impose retail prices. During operation, the agreement is assumed to have an impact on retailers and production consumers, and the impact of the value β on revenue can be calculated.

[0075] As a preferred technical solution, the energy router decides its own transactions based on changes in risk and operating costs.

[0076]

[0077]

[0078]

[0079]

[0080]

[0081]

[0082]

[0083]

[0084] in and is the operating cost with or without energy trading; and is the decision variable for including or not including operational transactions, α i Reflects transaction capabilities during the transaction process; It is the cost of losses incurred by producers and sellers during loading and unloading; is the producer's own storage capacity, It is the producer’s own unloading capacity; It is the capacity of the producer per unit time; is the load per unit time of producers and sellers; and are the values ​​transmitted and received by the main grid, and the values ​​charged and discharged by the producer and seller, respectively.

[0085]

[0086] The formula takes into account prices and multiple situations. The response of production consumers is modeled by a lower-level problem model. The objective function of producers and sellers as followers has two parts: the first term represents the internal cost function of production consumers without transaction items, and the second term represents the cost of transaction items with transaction costs, which is equal to the sum of internal cost functions and payments made by other producers and sellers. The function includes ET costs with distribution networks and batteries. The difference between these two expressions lies in whether transactions with storage power stations are directly considered, which ensures the power balance of energy routers in each unit time and in each production and sales mode, and ensures that the sum of export capabilities of production consumers with surplus energy is equal to the sum of import capabilities of surplus production consumers. Energy transactions between distribution networks and production consumers are restricted, and restrictions on energy transmission and reception between distribution networks and producers and sellers are defined, thereby reflecting that different transaction capabilities in the transaction process are less than their production capabilities.

[0087] Compared with the prior art, the present invention has the following beneficial effects:

[0088] 1. The present invention achieves energy conservation and efficiency improvement by utilizing the characteristics of bidirectional energy flow, information monitoring and remote operation, thereby increasing its flexibility and scalability. In addition, the invention takes into account the impact of uncertain factors such as the output of wind turbines and photovoltaics in microgrids on energy supply, and implements a reasonable collaborative optimization operation strategy to ensure orderly and reasonable energy interaction of energy routers in the context of uncertain wind and solar energy output, maximize social profits, and ensure fair income distribution.

[0089] 2. The present invention helps select the path with the lowest cost and achieves the lowest loss transmission by exhaustively searching and comparing all feasible transmission paths, so as to achieve the minimum loss of multiple energy flows. In order to achieve congestion-free transmission, two priority sorting methods are proposed to determine the energy transmission priority of multiple transactions based on the quotation and transaction volume. However, the centralized energy routing algorithm places all the computational burden on the network control center, which makes the network control center have extremely high computing power. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] Figure 1 A schematic diagram of the structure of multiple microgrids with energy routers according to the present invention;

[0091] Figure 2 A schematic diagram depicting the impact of the balance between expected profit and profit variability (measured by CVaR) of the present invention;

[0092] Figure 3 Schematic diagram of comparative analysis of Cvar and Var of the present invention;

[0093] Figure 4 Schematic diagram of production capacity parameters of the producer and seller of the present invention;

[0094] Figure 5 It is a schematic diagram of the time division table and real-time charging for each time period of the present invention. DETAILED DESCRIPTION

[0095] The following describes embodiments of the present invention in detail, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. In the description of the present invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," and the like, indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore should not be construed as limiting the present invention. In addition, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features indicated. Thus, features defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, "multiple" means two or more, and "several" means one or more, unless otherwise specifically defined.

[0096] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed, detachable, or integral connections; mechanical, electrical, or mutually communicative connections; direct or indirect connections through an intermediate medium; and internal communication between two components or interaction between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0097] In the present invention, unless otherwise expressly specified or limited, a first feature being "above" or "below" a second feature may include the first and second features being in direct contact, or may include the first and second features being in contact not directly but through another feature between them. Furthermore, a first feature being "above," "above," and "above" a second feature may include the first feature being directly above or obliquely above the second feature, or may simply mean that the first feature is higher in level than the second feature. A first feature being "below," "below," and "below" a second feature may include the first feature being directly below or obliquely below the second feature, or may simply mean that the first feature is lower in level than the second feature.

[0098] The disclosure below provides many different embodiments or examples for realizing different structures of the present invention. In order to simplify the disclosure of the present invention, the components and settings of specific examples are described below. Of course, they are merely examples and are not intended to limit the present invention. In addition, the present invention may repeat reference numbers and / or reference letters in different examples. Such repetition is for the purpose of simplicity and clarity and does not in itself indicate the relationship between the various embodiments and / or settings discussed. In addition, the present invention provides examples of various specific processes and materials, but those skilled in the art will recognize the application of other processes and / or the use of other materials.

[0099] In order to better understand the above technical solution, the above technical solution will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0100] According to the attached Figure 1-5 As shown, an embodiment of the present invention provides a multi-microgrid bidding energy management method based on an energy router, comprising the following steps:

[0101] Firstly, a multi-microgrid structure including energy routers is proposed. It regards multiple MGERs as a multi-agent system composed of intelligent individuals composed of energy routers. The energy router is considered as a public information and energy flow platform that shares its activity information with the distribution network center. Then, based on this, an integrated energy microgrid optimization operation strategy based on conditional value at risk (CVaR) and Nash negotiation model is proposed.

[0102] A CVaR-based master-slave profit allocation model is then established to measure the risk of wind and solar uncertainty. The contribution of microgrids to social welfare is measured by the degree of MGER transaction cost loss. Each microgrid conducts asymmetric negotiations based on the scale of its social profit contribution to achieve a fair distribution of cooperative surplus.

[0103] The effectiveness of the proposed method is verified through case analysis. Simulation results show that the proposed method can maximize the social profit of MGER in a multi-microgrid structure. Based on the cooperative benefits of the microgrid alliance, MGER can achieve fair energy distribution in terms of social profit contribution. This provides a feasible reference scheme for the operation decision-making of energy router multi-microgrid systems between economic benefits and risk levels.

[0104] The interaction pattern between the energy microgrid and the distribution network center is represented as a Stackelberg game model. As a leader, the distribution network center provides pricing decisions, enabling the energy microgrid to participate in energy management and achieve maximum social profit. As an energy router that tracks the changes in the distribution network center's decisions, it has only one "black box" of output and use. Through cooperation in the distribution network center, energy configuration is shared. As a leader, the distribution network center needs to make a choice between expected profit and CVaR to maximize social profit. Multiple microgrid structures with energy routers;

[0105] The distribution network center will incorporate the energy routers of all participants into the "black box" of the "power group" for unified management to achieve optimal energy supply between the interconnected multi-energy systems (MES). In addition, the uncertainty brought by potential risks, energy routers, microgrids, wind power and solar power generation, in order to improve the practicality of the model, and the uncertainty brought by risk factors are taken into account in the distribution network center. The interconnected multi-energy system (MES) can control the overall risk level from a holistic perspective;

[0106] As a leader, the distribution network coordinates the supply and demand of the microgrid. When the microgrid has energy procurement needs, the distribution network needs to provide energy. When the microgrid generates too much energy, the distribution network needs to purchase electricity to increase the penetration rate of new energy in the microgrid itself. When there is a mismatch within the microgrid, it requires retailers to coordinate with the distribution network. Generally speaking, the sales price is lower than the purchase price to encourage production and consumers to arrange internal energy. However, during the coordination process, the microgrid itself cannot obtain energy unlimitedly during low-price periods, and the distribution network itself is not an infinitely functional system, which may lead to the collapse of the power grid. Therefore, in order to achieve coordinated management and reduce the impact on the distribution network (DN), the goal of retailers is to control the risk level and cope with uncertainty while maximizing social profits.

[0107] As an energy router microgrid that changes with the response of the distribution network, the energy router participates in the production and distribution of energy. Therefore, the energy router is called a producer and seller. As a follower that changes with the response of the distribution network, the energy router determines its own transactions based on changes in risks and operating costs;

[0108] MES is jointly established by multiple microgrids containing ER. When configuring, the distribution center only needs to consider the total configuration cost of MES and the risk operation cost brought by CVAR calculation. The distribution center needs to reasonably configure MES to minimize the operating cost of the system. At the same time, each microgrid containing ER will adjust the output of various devices in the system according to the configuration of the distribution center to flatten the net load. Therefore, this problem is a two-level optimization problem. This paper requires the use of a solver or heuristic algorithm to solve the problem of maximizing the overall social profit of multiple microgrids. The microgrid model and the distribution network model are non-convex linear models. In order to facilitate the solution, in the non-convex problem of joint optimization of multiple energy microgrids, these nonlinear constraints are relaxed and the problem is decomposed into two sub-problems to solve, namely solving the global cost minimization sub-problem (P1) and the energy greed acquisition sub-problem (P2) tracking the distribution company to complete market settlement, so that the problem can be solved by the solver;

[0109] By relaxing the nonlinear terms, problems in local optimality can be transformed into global optimality. Additional augmented variables need to be introduced for expansion. Then, the spatial branch and bound method can be used. This is the most general algorithm for solving mixed integer nonlinear programming (MINLP). It is different from the traditional branch and bound method for solving mixed integer linear programming (MILP) problems. Relaxation is no longer a simple linear relaxation, but an approximate function that uses a convex underestimater or concave overestimater for each constraint and optimization objective, providing lower / double bound pruning. The branch and bound method not only needs to reduce the upper and lower bounds of the integer variables, but also needs to restrict the variables included in the constraints and update the corresponding approximate functions. In the process of relaxing variables, the constraints on energy trading are mainly discussed. In the energy trading process, due to the mutual exclusion between the same buyer and seller, they cannot occur simultaneously;

[0110] The complementary relaxation condition of the lower energy level model can be expressed as:

[0111]

[0112]

[0113]

[0114]

[0115]

[0116]

[0117]

[0118]

[0119]

[0120]

[0121] At the optimal solution point of the lower model, we have

[0122]

[0123]

[0124]

[0125]

[0126] The original two-level collaborative planning model can be transformed into a single-level MPEC problem, making it a non-convex model that is not suitable for direct solution. In this section, we will use the Big-M method to apply convex constraints to the non-convex part of the above model and linearize it using strong duality theory;

[0127] Using the Big-M method, by introducing several 0-1 variables, the original nonlinear constraint is equivalently transformed into a mixed integer linear constraint. This paper discusses the constraint problem of ET. In the process of ET, it is impossible to sell and buy energy for the same buyer and seller at the same time. Therefore, no matter whether the energy is increasing or decreasing, there is a formula transmission and distribution model;

[0128]

[0129]

[0130]

[0131]

[0132] It is the energy that producers and sellers sell to the distribution network; is the energy purchased by producers and sellers from the distribution network; M is a sufficiently large constant;

[0133] Based on the strong duality theory, the bilinear terms can be linearized, and the previous model can be transformed into a mixed integer second-order cone programming (MISOCP) model that can be optimized and solved. The objective function is shown as follows:

[0134]

[0135] Commercial solvers are available for this convex optimization problem and have been market approved;

[0136] Perform scenario generation and restoration, followed by CVAR calculation. Cornish-Fisher is a continuous asymptotic expansion method for the quantile of a standardized random variable. It expands the quantile function of a real distribution into a polynomial function of the quantile function of a standard normal distribution. The first four orders of expansion are expressed as follows:

[0137]

[0138] Where z α is the quantile α of the standard normal distribution, κ3 and κ4 represent the skewness and kurtosis respectively. Under this quantile expansion, the loss random variable can be expressed as:

[0139]

[0140] Where μ l and σ l They represent the expected value and standard deviation of the loss random variable in period l, respectively. It can be seen that without knowing the true distribution of the loss random variable, the skewness κ3 and kurtosis κ4 are estimated from the sample, and the asymptotic expansion and value of the quantile are obtained;

[0141] Considering the potential uncertainties in multiple aspects of energy supply, consumption, and transaction, which increase the risk of distribution network operation, I represents the number of microgrids. The risk of multiple uncertainties in the real-time stage can be expressed as Equation (44):

[0142]

[0143] Among them, s1, s2, ..., sN are sample data, α is the confidence level, and η i,w ,η i,v Calculate CVaR using auxiliary variables related to consumer i and scenario w;

[0144] In order to avoid false regression and eliminate heteroscedasticity, it is usually necessary to take the logarithmic wind and solar series time attributes of the data without changing the attributes and correlation of the time series. In order to obtain smooth data and make the data closer to normal distribution, the formula for taking the series is:

[0145] rt=-log(p)+log(p)

[0146] Estimate the failure percentage at a 95% confidence level as Figure 3 As shown;

[0147] It can be seen that most r are smaller than the corresponding VR and CVaR estimates, that is, the risk estimates are essentially valid. If r is greater than the corresponding VaR or CVaR value, the risk estimation fails. Obviously, CVaR is more reliable for the data. Then, compared with the theoretical value and the CVaR method, the VaR method is obviously not suitable for risk measurement in the power bidding market. The CVaR method estimates market risk well, while the VaR method has significant differences from the actual situation when estimating risk.

[0148] Case Analysis

[0149] This article uses Figure 1 The three-node model in the paper gives the parameters of the three microgrids for arithmetic calculations. The parameters of these devices are set according to Peng et al. (2021) and Na et al. (2023). The failure percentage at the 95% confidence level is estimated in Table 1. The parameters of each device are shown in Table 2. The simulation uses a computer configuration of I7-12700, RTX306012G and MATLAB2021b. The producer and load tariffs are gradually adjusted as shown in the following example. Figure 4 As shown;

[0150] Producer and load tariffs such as Figure 5 As shown;

[0151] CVaR Impact Analysis

[0152] Calculate the CVaR and expected profit at the 95% confidence level for different risk factors by modifying the parameter in the equation, β. The effect of this parameter on the balance between expected profit and profit variability (measured by CVaR) is described as follows: Figure 2 As shown;

[0153] The expected profits and CVaR of producers and sellers with different β values ​​should be listed as: (a) expected profits and CVaR1 of producers and sellers; (b) expected profits and CVaR2 of producers and sellers; (c) expected profits and CVaR3 of the market. The first point is that by solving the parameter β close to 0, which represents the maximum profit, producers and sellers may be risk-averse and approximate the optimal trading strategy of producers and sellers when they do not consider CVaR. It can be seen that as the bidding degree increases, producers and sellers become more and more risk-averse, and the expected returns decrease accordingly. The distribution network, as a leader, takes the maximization of overall profits as the overall goal, and analyzes the impact of different risk indicators on the trading profits of each producer and seller, thereby guiding the output of producers and sellers of new energy transactions and promoting the amount of new energy consumption;

[0154] Simulation results analysis

[0155] The transaction price between the distribution network and buyers and sellers depends on their attitude towards risk, which is caused by the uncertainty of wind and solar energy output. Retailers determine reasonable prices within a price range by setting minimum and maximum values ​​to maximize social welfare. Therefore, risk-seeking retailers may offer competitive prices to attract prosumers to participate in energy sharing while taking on relatively high risks.

[0156] Iterative Analysis

[0157] The model is solved using the CPlex solver. The optimal risk transaction value of each producer and seller is obtained through sorting and looping algorithms, and the optimal price curve that maximizes overall welfare is obtained. The method converges after 86 iterations, with a computation time of 689 seconds. After meeting the iterative convergence result, the residual error of each microgrid iteration converges to within 10-3. Therefore, the method proposed in this paper has good convergence performance and computational efficiency, meeting the needs of optimized scheduling.

[0158] Negative prices represent the energy selling prices of producers and sellers, while positive prices represent the purchase prices of energy. For profit, when distribution network operators sell energy to producers and sellers, the sales price is often higher than the purchase price. However, due to the load price limit, the purchase price cannot be higher than the load price, making it profitable for producers and sellers. Positive prices represent the purchase quantity of producers and sellers, while negative prices represent the sales quantity.

[0159] Draw conclusions

[0160] This paper proposes a two-layer energy interaction framework for multiple microgrids with energy routers, which can help the distribution network center coordinate energy transactions with energy routers. The multi-microgrid containing energy routers is regarded as a multi-agent system, and its activity information can be shared with the distribution network center. On this basis, we propose a static game theory model with the distribution network center as the leader and the microgrid containing energy routers as the follower. At the same time, risk measurement methods - VaR and CVaR calculations - are introduced to identify and quantify potential losses under the uncertainty of wind and solar output. The leader makes a balanced choice between income and risk, determines price conditions under the current risk indicators, encourages followers to engage in energy transactions, expands wind and solar energy consumption, and improves the overall return of the microgrid group. Considering the wind and solar power generation proposed in this paper In order to deal with the uncertainty of electricity, in the multi-microgrid bidding strategy using energy routers, energy management is divided into two stages. In the first stage, a master-slave profit distribution model based on the conditional risk value is established based on the interest needs of each microgrid and the energy supply and demand relationship, realizing the risk measurement of the dual uncertainty of wind and solar energy. In the second stage, the above problem is decomposed into the minimum cost sub-problem and the maximum transaction quantum problem of asymmetric negotiation, and energy is fairly distributed according to the contribution value of each microgrid containing energy routers. The simulation results show that the model enhances the ability of multiple microgrid systems with energy routers to cope with uncertain wind and solar energy output, improves the effectiveness and rationality of energy interaction between microgrid energy routers and supporting networks, and maximizes the profits of multiple microgrids, ensuring fair distribution of stakeholders.

[0161] Future work will include analyzing and formulating optimization problems for microgrids with energy routers in different configurations, further considering classification studies of energy routers in microgrids, and determining the impact of different microgrids on multiple microgrid systems to improve the widespread and universal application of energy routers in multiple microgrid systems.

[0162] As a preferred embodiment of this invention, this strategy aims to achieve the effectiveness and rationality of the energy interaction between MGER and the distribution network, while maximizing social welfare and ensuring the fairness of income distribution.

[0163] As a preferred embodiment of this invention, in order to clearly describe the coupling relationship of various energy sources in energy routers under multiple microgrids, a matrix method is used to describe the mutual relationship and interaction mode of electric energy in the energy router from the perspective of port devices. By connecting the input, output and conversion of electric energy together and imposing constraints to simplify the calculation of the distribution network center, it can simultaneously meet the energy supply and distribution needs of microgrids in different periods.

[0164] As a preferred embodiment of this invention, in order to clearly describe the coupling relationship of various energy sources in energy routers under multiple microgrids, a matrix method is used to describe the mutual relationship and interaction mode of electric energy in the energy router from the perspective of port devices. By connecting the input, output and conversion of electric energy together and imposing constraints to simplify the calculation of the distribution network center, it can simultaneously meet the energy supply and distribution needs of microgrids in different periods.

[0165] As a preferred embodiment of this invention, in order to clearly describe the coupling relationship of various energy sources in energy routers under multiple microgrids, a matrix method is used to describe the mutual relationship and interaction mode of electric energy in the energy router from the perspective of port devices. By connecting the input, output and conversion of electric energy together and imposing constraints to simplify the calculation of the distribution network center, it can simultaneously meet the energy supply and distribution needs of microgrids in different periods.

[0166] As a preferred embodiment of the present invention, the energy conversion matrix model is shown as the equation Z=XI Where Z, X, and I represent the port output power matrix, conversion coupling matrix, and port input power matrix of the energy router. In the conversion coupling matrix X, the diagonal X nn Indicates the network loss coefficient and energy distribution coefficient in the conversion, the diagonal X mn It represents the energy conversion coefficient s between device m and device n. The conversion coefficient includes the energy distribution coefficient and the energy efficiency coefficient. a represents the energy distribution coefficient. If the sum of the coefficients of the devices at the same port is 1, b is the port energy conversion efficiency, and c is the energy transmission loss coefficient.

[0167] As a preferred embodiment of this invention, the decision model of the distribution network is as follows: Where i represents the measure of producers and sellers; t represents the measure of time; N represents the total number of producers and sellers; T represents the sum of all time; Ω represents the total capacity of producers and sellers; π i represents the probability that producers and sellers are willing to take risks when purchasing electricity, β is the weighted coefficient of the profit given up by producers and sellers when load is reduced, and ζ i Auxiliary calculation of the value at risk (VaR) associated with the producer-seller, γ and the constraints associated with the Lagrange multiplier operator; η i,w ,η i,v Wind and sunlight condition values ​​and risk indicators, the price of energy purchased by producers and sellers from the distribution grid; the price at which producers and distributors sell energy to the distribution network, the price at which producers and distributors sell energy to the distribution network, Producers and sellers sell energy to the distribution grid, Producers and sellers purchase energy from the distribution network. The upper and lower limits on energy sales by producers and sellers in the distribution network, Floors and floors for purchasing energy from producers and sellers in the distribution network New York, the average price at which energy is sold by producers and distributors to consumers; the average price at which producers and distributors sell energy to the distribution network;

[0168] Expected retailer revenue (expected sales revenue to production consumers minus the expected cost of purchasing from DN), the trade-off between expected revenue and risk (CVaR multiplied by the weighting parameter β), the larger the β value, the greater the preference for risk aversion, and a lower β value is chosen to obtain higher revenue to reduce expected costs. In contrast, conservative retailers tend to choose a larger β value to increase the weight of risk aversion, and limit the transaction price offered to consumers to a time interval. In order to limit the market power of retailers, a daily average price is set to impose retail prices. During operation, the agreement is assumed to have an impact on retailers and production consumers, and the impact of the value β on revenue can be calculated.

[0169] As a preferred embodiment of this invention, the energy router decides its own transaction based on the changes in risk and operating costs.

[0170]

[0171]

[0172]

[0173]

[0174]

[0175]

[0176]

[0177]

[0178] in and is the operating cost with or without energy trading; and is the decision variable for including or not including operational transactions, α i Reflects transaction capabilities during the transaction process; It is the cost of losses incurred by producers and sellers during loading and unloading; is the producer's own storage capacity, It is the producer’s own unloading capacity; It is the capacity of the producer per unit time; is the load per unit time of producers and sellers; and are the values ​​transmitted and received by the main grid, and the values ​​charged and discharged by the producer and seller, respectively.

[0179]

[0180] The formula takes into account prices and multiple situations. The response of production consumers is modeled by a lower-level problem model. The objective function of producers and sellers as followers has two parts: the first term represents the internal cost function of production consumers without transaction items, and the second term represents the cost of transaction items with transaction costs, which is equal to the sum of internal cost functions and payments made by other producers and sellers. The function includes ET costs with distribution networks and batteries. The difference between these two expressions lies in whether transactions with storage power stations are directly considered, which ensures the power balance of energy routers in each unit time and in each production and sales mode, and ensures that the sum of export capabilities of production consumers with surplus energy is equal to the sum of import capabilities of surplus production consumers. Energy transactions between distribution networks and production consumers are restricted, and restrictions on energy transmission and reception between distribution networks and producers and sellers are defined, thereby reflecting that different transaction capabilities in the transaction process are less than their production capabilities.

[0181] Throughout this specification, reference to terms such as "an embodiment," "one embodiment," "certain embodiments," "illustrative embodiments," "example," "specific example," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0182] The basic principles, main features, and advantages of the present invention have been described above in conjunction with specific embodiments. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely preferred examples of the present invention and are not intended to limit the present invention. Various changes and improvements may be made to the present invention without departing from the spirit and scope of the present invention, and such changes and improvements fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A multi-microgrid bidding energy management method based on energy router, characterized in that: The steps include: Firstly, a multi-microgrid structure including energy routers is proposed, which regards multi-MGER as a multi-agent system composed of intelligent individuals composed of energy routers. The energy router is considered as a public information and energy flow platform that shares its activity information with the distribution network center. Then, based on this, an integrated energy microgrid optimization operation strategy based on conditional value at risk (CVaR) and Nash negotiation model is proposed. Then, a CVaR-based master-slave distribution company profit model is established to measure the risk of wind and solar uncertainty. The contribution of microgrids to social welfare is measured by the degree of MGER transaction cost loss. Each microgrid conducts asymmetric negotiations based on the scale of social profit contribution to achieve a fair distribution of cooperative surplus. The effectiveness of the method is verified by case analysis. The simulation results show that the method can maximize the social profit of MGER in a multi-microgrid structure. Based on the cooperative benefits of the microgrid alliance, MGER can achieve fair energy distribution in terms of social profit contribution, which provides a feasible reference solution for the operation decision of the energy router multi-microgrid system between economic benefits and risk levels. The interaction mode between energy microgrid and distribution network center is represented as Stackelberg game model. As a leader, the distribution network center provides pricing decisions to enable energy microgrid to participate in energy management and achieve the maximum social profit. As an energy router that tracks the changes in the distribution network center's decisions, it has only one output and a "black box" used. Through cooperation in the distribution network center, energy configuration is shared. As a leader, the distribution network center needs to make a choice between expected profit and CVaR to maximize social profit. Multiple microgrid structures with energy routers; The distribution network center will incorporate the energy routers of all participants into the "black box" of the "power group" for unified management to achieve the best energy supply between the interconnected multi-energy systems (MES). In addition, the uncertainty caused by potential risks, energy routers, microgrids, wind power and solar power generation, in order to improve the practicality of the model, and the uncertainty caused by risk factors are considered in the distribution network center. The interconnected multi-energy system (MES) can control the overall risk level from a holistic perspective; As a leader, the distribution network coordinates the supply and demand of the microgrid. When the microgrid has energy procurement needs, the distribution network needs to provide energy, and when the microgrid generates too much energy, the distribution network needs to purchase electricity to increase the penetration rate of new energy in the microgrid itself. When there is a mismatch within the microgrid, it requires retailers to coordinate with the distribution network. Generally speaking, the sales price is lower than the purchase price, which encourages production and consumption to arrange internal energy. However, in the coordination process, the microgrid itself cannot obtain energy unlimitedly during low-price periods, and the distribution network itself is not an unlimited functional system, which may lead to the collapse of the power grid. Therefore, in order to achieve coordinated management and reduce the impact on the distribution network (DN), the goal of retailers is to control the risk level and deal with uncertainty while maximizing social profits; As an energy router microgrid that changes with the response of the distribution network, the energy router participates in the production and distribution of energy. Therefore, the energy router is called a producer and seller. As a follower that changes with the response of the distribution network, the energy router determines its own transactions based on the changes in risks and operating costs; MES is jointly established by multiple microgrids including ER. When configuring, the distribution center only needs to consider the total configuration cost of MES and the risk operation cost brought by CVAR calculation. The distribution center needs to reasonably configure MES to minimize the operating cost of the system. At the same time, each microgrid including ER will adjust the output of various equipment in the system according to the configuration of the distribution center to flatten the net load. Therefore, this problem belongs to a two-level optimization problem. This paper requires the use of a solver or heuristic algorithm to solve the problem of maximizing the overall social profit of multiple microgrids. The microgrid model and the distribution network model are non-convex linear models. In order to facilitate the solution, in the non-convex problem of joint optimization of multi-energy microgrids, these nonlinear constraints are relaxed and the problem is decomposed into two sub-problems to solve, namely, solving the global cost minimization sub-problem (P1) and the energy greedy acquisition sub-problem (P2) tracking the distribution company to complete the market clearing, so that the problem can be solved by the solver; By relaxing the nonlinear terms, the problem in the local optimum can be transformed into the global optimum. Additional enhanced variables need to be introduced for expansion. Then the spatial branch and bound method can be used. This is the most general algorithm for solving mixed integer nonlinear programming (MINLP). It is different from the traditional branch and bound method for solving mixed integer linear programming (MILP) problems. The relaxation is no longer a simple linear relaxation, but an approximate function that uses a convex underestimater or concave overestimater in each constraint and optimization target, providing lower / double bound pruning. The branch and bound method not only needs to reduce the upper and lower bounds of integer variables, but also needs to restrict the variables included in the constraints and update the corresponding approximate functions. In the process of relaxing variables, the constraints on energy trading are mainly discussed. In the process of energy trading, due to the mutual exclusion between the same buyer and seller, they cannot occur at the same time; The complementary relaxation condition of the lower energy level model can be expressed as: At the optimal solution point of the lower model, we have ie; The original two-level collaborative planning model can be transformed into a single-level MPEC problem, making it a non-convex model that is not suitable for direct solution. In this section, we will use the Big-M method to apply convex constraints to the non-convex part of the above model and linearize it using strong duality theory; Using the Big-M method, by introducing several 0-1 variables, the original nonlinear constraint is equivalently transformed into a mixed integer linear constraint. This paper discusses the constraint problem of ET. In the process of ET, it is impossible to sell energy and buy energy for the same buyer and seller at the same time. Therefore, no matter whether the energy is increasing or decreasing, there is a formula for the transmission and distribution model; It is the energy that producers and sellers sell to the distribution network; It is the energy that producers and sellers purchase from the distribution network; M is a sufficiently large constant; Based on the strong duality theory, the bilinear terms can be linearized, and the previous model can be transformed into a mixed integer second-order cone programming (MISOCP) model that can be optimized and solved. The objective function is shown as follows: Commercial solvers are available for this convex optimization problem and are market-approved; Perform scenario generation and restoration, followed by CVAR calculation. Cornish-Fisher is a continuous asymptotic expansion method for the quantile of a standardized random variable. It expands the quantile function of a real distribution into a polynomial function of the quantile function of a standard normal distribution. The first 4 orders of expansion are expressed as follows: Where z α is the quantile α of the standard normal distribution, κ3 and κ4 represent the skewness and kurtosis respectively. Under this quantile expansion, the loss random variable can be expressed as: In the formula, μ l and σ l They represent the expected value and standard deviation of the loss random variable in period l respectively. It can be seen that without knowing the true distribution of the loss random variable, the skewness κ3 and kurtosis κ4 are estimated from the sample, and the asymptotic expansion and value of the quantile are obtained; Considering the potential uncertainties in multiple aspects of energy supply, consumption, and trading, the risk of distribution network operation is increased. I represents the number of microgrids. The risk of multiple uncertainties in the real-time stage can be expressed as Equation (44): Among them, s1, s2, ..., sN are sample data, α is the confidence level, and η i,w , η i,v Calculate CVaR using auxiliary variables associated with consumer i and scenario w; In order to avoid false regression and eliminate heteroscedasticity, it is usually necessary to take the logarithmic wind and solar series time attributes of the data without changing the attributes and correlation of the time series. In order to obtain smooth data and make the data closer to normal distribution, the formula for taking the series is: rt=-log(p)+log(p) It is known that most r are smaller than the corresponding VR and CVaR estimates, that is, the risk estimates are essentially valid. If r is larger than the corresponding VaR or CVaR value, the risk estimation fails. Obviously, CVaR is more reliable for the data. Then, compared with the theoretical value and the CVaR method, the VaR method is obviously not suitable for risk measurement in the power bidding market. The CVaR method estimates market risk well, while the VaR method has significant differences from the actual situation when estimating risk. Example Analysis This paper uses the three-node model in Figure 1 and gives the parameters of three microgrids for arithmetic calculations. The parameters of these devices are set according to Peng et al. (2021) and Na et al. (2023); CVaR Impact Analysis Calculate the CVaR and expected profit at 95% confidence level for different risk factors by modifying the parameters in the equation, β; The expected profits and CVaR of producers and sellers with different β values ​​should be listed as: (a) expected profits and CVaR1 of producers and sellers; (b) expected profits and CVaR2 of producers and sellers; (c) expected profits and CVaR3 of the market. The first point is that by solving the parameter β close to 0, which represents the maximum profit, producers and sellers may be risk-averse and approximate the optimal trading strategy of producers and sellers when they do not consider CVaR. It can be seen that as the bidding degree increases, producers and sellers become more and more risk-averse, and the expected return decreases accordingly. The distribution network, as a leader, takes the maximization of overall profit as the overall goal. The impact of different risk indicators on the trading profit of each producer and seller is analyzed, thereby guiding the output of producers and sellers of new energy transactions and promoting the amount of new energy consumption; Simulation results analysis The transaction price between the distribution network and buyers and sellers depends on the attitude of buyers and sellers towards risk, which is caused by the uncertainty of wind and solar output. Retailers determine reasonable prices within a price range by setting minimum and maximum values ​​to maximize social welfare. Therefore, risk-seeking retailers may offer competitive prices to attract production consumers to participate in energy sharing while taking relatively high risks. Iterative analysis The model is solved by using the cplex solver, and the optimal risk transaction value of each producer and seller is obtained through sorting and loop algorithms to obtain the optimal price curve that maximizes the overall welfare. This method converges after 86 iterations, and the calculation time is 689 seconds. After meeting the iterative convergence result, the residual of each microgrid iteration converges to within 10-3. Therefore, the method proposed in this paper has good convergence performance and calculation efficiency, which meets the needs of optimal scheduling. Negative prices represent the energy selling price of producers and sellers, while positive prices represent the purchase price of energy. For profit, when the distribution network operator sells energy to producers and sellers, the sales price is often higher than the purchase price. Limited by the load price, the purchase price cannot be higher than the load price, making it profitable for producers and sellers. Positive prices represent the purchase quantity of producers and sellers, while negative prices represent the sales quantity. Draw Conclusion This paper proposes a two-layer energy interaction framework for multi-microgrids with energy routers, which can help the distribution network center coordinate energy transactions with energy routers. The multi-microgrids containing energy routers are regarded as multi-agent systems, and their activity information can be shared with the distribution network center. On this basis, we propose a static game theory model with the distribution network center as the leader and the microgrids containing energy routers as followers. At the same time, risk measurement methods - VaR and CVaR calculations - are introduced to identify and quantify potential losses under the uncertainty of wind and solar output. Leaders make a balanced choice between income and risk, determine price conditions under the current risk indicators, encourage followers to engage in energy trading, expand wind and solar consumption, and improve the overall return of the microgrid group. Considering the wind and solar power generation proposed in this paper In the multi-microgrid bidding strategy using energy routers, energy management is divided into two stages. In the first stage, a master-slave model based on the interest needs of each microgrid and the energy supply and demand relationship is established to distribute the company's profit based on the conditional risk value, and the risk measurement of the dual uncertainty of wind and solar energy is realized. In the second stage, the above problem is decomposed into the minimum cost sub-problem and the maximum transaction quantum problem of asymmetric negotiation, and energy is fairly distributed according to the contribution value of each microgrid containing energy routers. The simulation results show that the model enhances the ability of multiple microgrid systems with energy routers to cope with uncertain wind and solar energy output, improves the effectiveness and rationality of energy interaction between microgrid energy routers and supporting networks, and maximizes the profits of multiple microgrids at the same time, ensuring fair distribution of interest participants; Future work will include analyzing and formulating optimization problems for microgrids with energy routers in different configurations, further considering classification studies of energy routers in microgrids, and determining the impact of different microgrids on multiple microgrid systems to improve the widespread and universal application of energy routers in multiple microgrid systems.

2. A multi-microgrid bidding energy management method based on energy router according to claim 1, characterized in that: The strategy aims to achieve the effectiveness and rationality of the energy interaction between MGER and the distribution network, while maximizing social welfare and ensuring fairness in income distribution.

3. The energy management method for multiple microgrids based on energy routers according to claim 1 is characterized by: In order to clearly describe the coupling relationship between various energy sources in energy routers under multiple microgrids, a matrix method is used to describe the mutual relationship and interaction mode of electric energy in energy routers from the perspective of port devices. By connecting the input, output and conversion of electric energy together and imposing constraints to simplify the calculation of the distribution network center, it can simultaneously meet the energy supply and distribution needs of microgrids in different periods.

4. The energy management method for multiple microgrids based on energy routers according to claim 1 is characterized by: The producers and sellers can cooperate with other producers and sellers to reduce operating costs. At the same time, producers and sellers maximize the use of the energy they can obtain from the distribution network under limited constraints. This can be called greedy. Therefore, we decompose the optimal scheduling problem into the minimum cost subproblem P1 and the maximum transaction quantum problem P2. The calculation formula is 5. The method for energy management based on energy routers for bidding of multiple microgrids according to claim 1 is characterized by: The transaction volume of the largest acquisition is: When the distribution network determines the supply and price, the market electricity price will be affected, which will affect the decision-making of electricity traders and consumers, and make appropriate negotiation decisions through formulas. The decision of the distribution network will affect the decision of the microgrid, and then affect the decision of the distribution network layer. This feedback loop can lead to adjustments in market prices and supply and demand, and ultimately tend to the optimal equilibrium state.

6. The method for energy management based on energy routers for bidding of multiple microgrids according to claim 1 is characterized by: The energy conversion matrix model is shown as Eq. Where Z, X, and I represent the port output power matrix, conversion coupling matrix, and port input power matrix of the energy router. In the conversion coupling matrix X, the diagonal X nn Represents the network loss coefficient and energy distribution coefficient in the conversion, and the diagonal X mn It represents the energy conversion coefficient s between device m and device n. The conversion coefficient includes the energy distribution coefficient and the energy efficiency coefficient. a represents the energy distribution coefficient. If the sum of the coefficients of the devices at the same port is 1, b is the port energy conversion efficiency and c is the energy transmission loss coefficient.

7. According to the energy router-based multi-microgrid bidding energy management method of claim 1, its characteristic is that: the decision model of the distribution network is as follows: in, i represents the measure of producers and sellers; t represents the measure of time; N represents the total number of producers and sellers; represents the sum of all time; Ω represents the total capacity of producers and sellers; π i represents the probability that producers and sellers are willing to take risks when purchasing electricity, β is the weighted coefficient of the producer and seller giving up profits when the load is reduced, ζ i Auxiliary calculation of the value at risk (VaR) associated with the producer-seller, γ constraints associated with the Lagrange multiplier operator; η i,w , η i,v Wind and sunlight condition values ​​and risk indicators, the price of energy purchased by producers and sellers from the distribution grid; the price at which producers and distributors sell energy to the distribution network, the price at which producers and distributors sell energy to the distribution network, Producers and sellers sell energy to the distribution grid, Producers and sellers purchase energy from the distribution network. The upper and lower limits on energy sales by producers and sellers in the distribution network, The upper and lower limits on energy purchases from producers and sellers in the distribution grid, New York, the average price at which energy is sold by producers and distributors to consumers; the average price at which energy is sold by producers and distributors to the distribution network; Expected retailer revenue (expected sales revenue to production consumers minus expected cost of purchases from DN), trade-off between expected revenue and risk (CVaR multiplied by weighting parameter β), larger β values ​​imply greater preference for risk aversion, lower β values ​​are chosen in order to obtain higher revenues at the expense of lower expected costs, in contrast, conservative retailers tend to choose larger β values ​​to increase the weight of risk aversion, limit transaction prices offered to consumers to time intervals, set daily average prices to impose retail prices in order to limit the market power of retailers, during operation, the protocol is assumed to have an impact on retailers and production consumers, and the impact of value β on revenue can be calculated.

8. The method for energy management based on energy routers for bidding of multiple microgrids according to claim 1 is characterized by: The energy router decides its own transactions based on changes in risk and operating costs. in and for operating costs with or without energy trading; and is the decision variable for including or not including operational transactions, α i Reflects the transaction ability during the transaction process; It is the cost lost by producers and sellers during loading and unloading; is the producer’s own storage capacity, It is the producer’s own unloading capacity; It is the capacity per unit time of the producer; is the load per unit time of producers and sellers; and are the values ​​of transmission and reception of the main grid, and the values ​​of charging and discharging of producers and sellers themselves, respectively. The formula takes into account prices and multiple situations. The response of production consumers is modeled by a lower-level problem model. The objective function of producers and sellers as followers has two parts: the first term represents the internal cost function of production consumers without transaction items, and the second term represents the cost of transaction items with transaction costs, which is equal to the sum of internal cost functions and other producers and sellers' payments. The function includes ET costs with distribution networks and batteries. The difference between these two expressions lies in whether transactions with storage power stations are directly considered, which ensures the power balance of energy routers in each unit time and in each production and sales mode, and ensures that the sum of export capacities of production consumers with surplus energy is equal to the sum of import capacities of surplus production consumers. Energy transactions between distribution networks and production consumers are restricted, and restrictions on energy transmission and reception between distribution networks and producers and sellers are defined, thereby reflecting the different transaction capacities in the transaction process that are less than their production capacities.

Images