Multi-agent game demand response method considering the interaction between SESS-MEA and EVA

Abstract

The present invention relates to a multi-agent game demand response method considering the interaction between SESS-MEA and EVA, comprising: a microgrid determines the response amount of different energy sources based on a first differentiated incentive rate and penalty price provided by SESS-MEA; the SESS-MEA considers the electric energy interaction with EVA according to the response amount of different energy sources to construct a first objective function; an EV user determines the required response amount of electric energy based on a second differentiated incentive rate and penalty price provided by EVA; the EVA considers the electric energy interaction with SESS-MEA according to the required response amount of electric energy to construct a second objective function; weights are assigned to the first objective function and the second objective function to obtain a multi-agent game model, and the demand response of the microgrid and EVs is obtained by solving the multi-agent game model. The present invention can reduce the impact of the microgrid on the stability of the power grid and improve the energy utilization efficiency between microgrids.

Description

Technical Field

[0001] The present invention relates to the technical field of electric energy dispatching, and more specifically to a multi-agent game demand response method considering the interaction between SESS-MEA and EVA. Background Art

[0002] With the increasing global attention to reducing greenhouse gas emissions and achieving sustainable development, the "dual carbon" goals, namely carbon peak and carbon neutrality, have become keywords for promoting economic and social development. Against this background, the development and utilization of renewable energy (RES) has developed rapidly. However, RES such as wind power and photovoltaics have obvious uncertainties and fluctuations, which bring challenges to the stability of the power grid, especially in terms of supply and demand balance.

[0003] In traditional power systems, supply-side regulation is the main way to maintain grid stability. However, with the increase in the proportion of RES and the improvement of grid stability requirements, relying solely on supply-side regulation can no longer meet the needs. Therefore, the importance of demand-side management has become increasingly prominent, among which demand response (DR) has been proposed and widely used as an effective demand-side management method. By incentivizing or guiding users to reduce electricity consumption during peak hours or increase electricity consumption during off-peak hours, DR can effectively smooth out grid load fluctuations and improve system operation efficiency.

[0004] As intermediaries for DR transactions, demand aggregators aggregate a large number of users' adjustable loads to participate in DR projects, becoming an important bridge between users and the power market. However, to further increase the penetration rate of RES and ensure the stable operation of the power grid, it is necessary to solve the problem of how to stimulate users' enthusiasm for participation while ensuring their benefits and the stability of the power grid.

[0005] The development of integrated energy systems and integrated demand response (IDR) provides new ideas for solving the above problems. Integrated energy aggregators (MEAs) are not limited to electricity, but also involve demand response aggregation of various energy forms such as gas and heat, providing a broader space for IDR transactions. Current research mainly focuses on how to build a reasonable trading model to balance the interests of different entities while considering the impact of energy coupling effects and price mechanisms. However, the response trading model established by the above research does not fully consider the uncertainty of demand response and the role of differentiated incentive rates, and ignores the impact of the interaction between many users and the power grid on the stability of the power grid.

[0006] At the same time, the rise of electric vehicles (EVs) has made electric vehicle aggregators (EVAs) another important DR transaction medium. EVAs participate in the DR market by aggregating the charging load of EVs, which helps to regulate the grid load, but at the same time, it is also necessary to consider the impact of differentiated incentive rates on the response costs of EV users.

[0007] Although existing research has made some progress, there are still some shortcomings: for example, the uncertainty of demand response, the role of differentiated incentive rates, and the impact of the interaction between many users and the grid on grid stability have not been fully considered. Summary of the invention

[0008] In view of this, in order to at least partially solve the above technical problems, the present invention provides a multi-agent game demand response method considering the interaction between SESS-MEA and EVA.

[0009] In order to achieve the above object, the present invention adopts the following technical solution:

[0010] A multi-agent game demand response method considering the interaction between SESS-MEA and EVA, comprising:

[0011] The microgrid determines the required response amounts of different energy sources based on the first differentiated incentive rate provided by the SESS-MEA; the SESS-MEA constructs a first objective function based on the penalty price and the required response amounts of the different energy sources and considering the electric energy interaction with the EVA;

[0012] The EV user determines the required response amount of electric energy based on the second differentiated incentive rate provided by the EVA; the EVA constructs a second objective function based on the required response amount of electric energy and taking into account the electric energy interaction with the SESS-MEA;

[0013] Weights are assigned to the first objective function and the second objective function to obtain a multi-agent game model, and the multi-agent game model is solved to obtain demand responses of the microgrid and EVs.

[0014] Preferably, the microgrid determines the required response amount of different energy sources with the goal of minimizing cost; the formula is expressed as:

[0015]

[0016] Where: The cost of purchasing natural gas for the microgrid; c g,t is the price of natural gas; and They are the operation and maintenance costs of the energy conversion device and the energy storage device respectively; and are the unit operation and maintenance costs of the energy conversion device and the energy storage device, respectively; is the incentive rate given by SESS-ME to the κth load of the ith energy in the yth microgrid at time t; is the total transmission power of the energy storage device; is the output power of the ζ-th energy conversion device; Power purchased from natural gas for microgrids; Cost of purchasing electricity from SESS-MEA for the microgrid.

[0017] Preferably, if the electric energy provided by the microgrid to SESS-MEA during the dispatch period is greater than the electric energy provided by SESS-MEA to the microgrid, SESS-MEA shall bear the cost of storing electric energy in the microgrid and shall not pay the microgrid for the electric energy; otherwise, the microgrid shall pay SESS-MEA for the excess electric energy. The formula is:

[0018]

[0019] Where: Provide electrical power to SESS-MEA for the microgrid; is the power given by SESS-MEA to the microgrid; τ is the price at which the microgrid purchases electricity from SESS-MEA.

[0020] Preferably, the first objective function is expressed as:

[0021]

[0022] Where: C MEA is the total benefit of SESS-MEA; C b Profits from SESS-MEA’s participation in demand response; to purchase costs for energy; is the equipment operation and maintenance cost; C re The cost of handling response uncertainty for energy storage devices; C f With C fa are the insufficient response penalty costs of microgrid and SESS-MEA, respectively.

[0023] Preferably, the present application uses carbon quotas to describe the interval length and determines the carbon trading cost through a two-stage tiered carbon trading mechanism, which is expressed as follows:

[0024]

[0025] E r =E a -E e -E CHP -E GB =E a -E p

[0026]

[0027] Where: C carbon is the carbon trading cost; e Carbon emission quota per unit of electricity for coal-fired units; is the electric power purchased by SESS-MEA from the grid; g,e With g,h Carbon emission quotas per unit of electricity and per unit of heat for gas-fired units; and are the output powers of GT, WHB and GB respectively; E e 、E CHP With E GB are the carbon emission quotas of coal-fired units, CHP and GB respectively; E p is the total carbon emission quota of SESS-MEA; E a With E r is the actual carbon emissions and carbon emissions trading volume; λ is the carbon trading base price; a is the initial interval range; b is the interval unit growth range; u is the transaction price growth range.

[0028] Preferably, the microgrid and SESS-MEA both include energy conversion devices and energy storage devices, and the energy conversion devices include: electric boilers, electric refrigerators, gas turbines, waste heat boilers, gas boilers, and absorption refrigerators.

[0029] Preferably, the EV user determines the required response amount of electric energy with the goal of minimizing cost, and the formula is expressed as:

[0030]

[0031] Where: C n,EV is the total cost of the nth EV; c EVA,t and are the price and power of electricity purchased by EV from EVA at time t respectively; and They are the power and reward for EVs to actively participate in reducing electricity consumption during peak hours, which is provided by EVA; is the incentive rate of the κth response provided by EVA to EV; is the discomfort cost of EV participating in demand response, which is proportional to the response amount and inversely proportional to the incentive rate. Since TL only causes the transfer of EV load, it has a small impact on EV, while IOL has a greater impact on EV. Therefore, in order to describe the impact of the two responses on the discomfort cost of EV, the impact weight coefficient ω is introduced κ ;c dr Discomfort cost per unit.

[0032] Preferably, the second objective function is expressed as:

[0033]

[0034] Where: C EVA is the total benefit of EVA; and The power of electricity purchased by EVA from the grid and the power of RES purchased from RES producers respectively; The price to buy RES for EVA.

[0035] Preferably,

[0036]

[0037] Where: P ME,t With P EM,t are the electric power transmitted from SESS-MEA to EVA and the electric power transmitted from EVA to SESS-MEA at time t; according to the energy consumption characteristics of SESS-MEA and EVA, the price c of electricity sold by SESS-MEA to EVA is set ME,t It is lower during 21:00-6:00, and during 11:00-14:00, the price at which EVA sells electricity to SESS-MEA is set c EM,t Lower.

[0038] Preferably, responding to multi-agent demands based on the multi-agent game model includes:

[0039] Solve the multi-agent game model to obtain the first target adjustment incentive rate, the target penalty price, and the second target adjustment incentive rate; and return them to the microgrid and the EVA user for repeated iterations until the iteration stop condition is met;

[0040] And adjust the incentive rate and penalty price according to the iterative target to determine the estimated response amount.

[0041] Preferably, the Steinberg game model is used to describe the response transaction process between multiple agents, and the multi-agent game model is solved using a particle swarm algorithm until the Nash solution of the model is obtained or the maximum number of iterations is reached.

[0042] The steps of adjusting the incentive rate and penalty price according to the iterated target and determining the estimated response volume include:

[0043] (1) SESS-MEA and MEA submit the demand response volume and incentive rate of microgrids and EVs to the trading market respectively;

[0044] (2) The trading market matches the information submitted by the responding demand party and the demand responding party, and publishes the clearing results.

[0045] It can be seen from the above technical solutions that the present invention discloses a multi-agent game demand response method considering the interaction between SESS-MEA and EVA.

[0046] First, in order to reduce the interaction between microgrids and grids and improve the energy utilization efficiency between microgrids, an IDR trading model between SESS-MEA and microgrids considering a two-stage tiered carbon trading mechanism is established. Among them, SESS-MEA not only provides differentiated incentive rates and penalty prices to microgrids as a response trading medium, but also acts as a shared energy storage station (SESS) to interact with each microgrid, and its energy storage device is used to deal with the uncertainty of the response.

[0047] Secondly, a DR transaction model between EVA and EV users is established, which uses incentive rate and response amount to describe the discomfort cost of EV users' response.

[0048] Finally, in order to improve the penetration rate of RES in the demand response trading model, a game model considering the interaction between SESS-MEA and EVA electricity was established.

[0049] Compared with the prior art, the present invention reduces the impact of microgrids on grid stability and improves energy utilization efficiency between microgrids by increasing the interaction between each microgrid and MEA; specifically, the beneficial effects of the present application are as follows:

[0050] 1) Increasing the interaction between each microgrid and MEA and the interaction between SESS-MEA and EVA is conducive to improving the economic efficiency of the model and the penetration rate of RES;

[0051] 2) As the weight coefficient q1 increases, the SESS-MEA benefits gradually decrease, while the EVA benefits increase. The model achieves the maximum benefit when (q1,q2) = (0.5,0.5), and the least amount of electricity purchased from the grid when (q1,q2) = (0.3,0.7). Therefore, according to the requirements of the model's economy and low carbon, the appropriate weight coefficient is selected.

[0052] 3) Compared with the traditional carbon trading mechanism and the tiered carbon trading mechanism, the two-stage tiered carbon trading mechanism that uses carbon quotas to describe the interval length has greater advantages in carbon emission reduction. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0054] Figure 1 A schematic diagram of the DR transaction process provided by the present invention;

[0055] Figure 2A flowchart for solving the game model provided by the present invention;

[0056] Figure 3 RES consumption for the four scenarios provided by the present invention;

[0057] Figure 4 This is the amount of electric energy interaction between SESS-MEA and EVA in scenario 4 provided by the present invention. DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] In order to fully tap the load flexibility potential, improve the penetration rate of renewable energy and the stability of the power grid, this embodiment proposes a multi-agent game demand response method considering the interaction between SESS-MEA (comprehensive energy aggregator) and EVA (electric vehicle aggregator).

[0060] Embodiment 1

[0061] This application first proposes a MEA with shared energy storage function (SESS-MEA), and establishes a comprehensive demand response trading model for the interaction between SESS-MEA and each microgrid, so as to reduce the impact of microgrids on grid stability and improve the responsiveness of microgrids.

[0062] In this model, MEA not only provides differentiated incentive rates and penalty prices to each microgrid, but also acts as a shared energy storage charging station (SESS) to interact with each microgrid. In addition, a two-stage tiered carbon trading mechanism using carbon quotas to describe the interval length is proposed to reduce the carbon emissions of this model.

[0063] In one embodiment, in order to reduce the instability of the power grid caused by the interaction between the microgrid and the power grid, and to improve the energy utilization efficiency of each microgrid, the microgrid constructed by the present invention interacts with the SESS-MEA for electric energy, and stores the electric energy in the energy storage device of the SESS-MEA. In order to effectively reduce carbon emissions, a two-stage stepped carbon trading mechanism using carbon quotas to describe intervals is introduced into the model. In addition, in order to increase the coupling between energy sources, the microgrid and the SESS-MEA model both include energy conversion devices and energy storage devices. Among them, the energy conversion device includes: an electric boiler (EB), an electric refrigerator (EC), a gas turbine (GT), a waste heat boiler (WHB), a gas boiler (GB), and an absorption refrigerator (AC), wherein the GT and WHB are combined to form a combined heat and power unit (CHP).

[0064] In another embodiment, the integrated demand response IDR model in the microgrid considers the demand response of four loads: electricity, gas, heat, and cooling, and the IDR types include on / off load (IOL), transferable load (TL), and emergency demand response load (EDRL). The IOL, TL, and EDRL models can be expressed as:

[0065]

[0066]

[0067] Where: It is the estimated response value of the κ-th type of response load of the ith energy load of the y-th microgrid at time t (i∈e,g,h,c;κ∈IOL,TL,EDRL). If the value is positive, it indicates that the microgrid load is reduced, otherwise the load increases. for The maximum value of . Formula (2) shows that the sum of TL in the scheduling period is 0.

[0068] The load of the microgrid after response can be expressed as:

[0069]

[0070] Where: P y,i,t and are the total loads of each energy before and after the microgrid responds, respectively.

[0071] Furthermore, the microgrid determines the required response of different energy sources with the goal of minimizing cost; that is, the microgrid determines the response amount based on the incentive rate provided by SESS-MEA to achieve the minimum cost. y,m It can be expressed as:

[0072]

[0073] Where: The cost of purchasing natural gas for the microgrid; c g,t is the price of natural gas; and They are the operation and maintenance costs of the energy conversion device and the energy storage device respectively; and are the unit operation and maintenance costs of the energy conversion device and the energy storage device, respectively; is the incentive rate given by SESS-ME to the κth load of the ith energy in the yth microgrid at time t; is the total transmission power of the energy storage device; is the output power of the ζ-th energy conversion device; power to purchase natural gas for the microgrid; Cost of purchasing electricity from SESS-MEA for the microgrid.

[0074] In this embodiment, when the microgrid and SESS-MEA exchange electric energy, if the electric energy provided by the microgrid to SESS-MEA during the scheduling period is greater than the electric energy provided by SESS-MEA to the microgrid, SESS-MEA bears the cost of storing electric energy in the microgrid but does not pay the cost of this part of electric energy to the microgrid; on the contrary, the microgrid needs to pay the cost of the excess electric energy to SESS-MEA. This energy purchase cost can be expressed as:

[0075]

[0076] Where: Provide electrical power to SESS-MEA for the microgrid; is the power given by SESS-MEA to the microgrid; τ is the price at which the microgrid purchases electricity from SESS-MEA.

[0077] Among them, the SESS-MEA model needs to meet the following constraints:

[0078] a) Power balance constraints

[0079]

[0080] Where: and They are wind and photovoltaic power generation power of microgrid respectively; and are the output and input power of the energy storage device (i∈g,h,c); is the energy coupling matrix of the energy conversion device, which represents the energy conversion relationship between the energy conversion devices; and is the input power and conversion efficiency of the energy conversion device; N is the total number of microgrids.

[0081] b) Equipment operation constraints

[0082]

[0083] Where: S y,i,t is the energy storage state of the energy storage device at time t (i∈g,h,c) and the total amount of electric energy stored in the SESS-MEA by the microgrid (i∈e); S y,i,min With S y,i,max The minimum and maximum capacities of the energy storage device and the minimum and maximum values ​​of the energy that the SESS-MEA allows the microgrid to store.

[0084] To further optimize the above technical solutions, SESS-MEA not only aggregates the responses of each microgrid, but also interacts with the microgrid for power. To improve the response enthusiasm of the microgrid, SESS-MEA gives each microgrid differentiated incentive rates and penalty prices, and uses energy storage devices to handle IDR uncertainty based on the opportunity constraint method.

[0085] In this embodiment, IDR uncertainty is modeled in the form of a random variable, and the uncertainty can be defined as ξ y,i,t ;

[0086]

[0087] Where: ΔP y,i,t is the actual response of the microgrid; y,i,t is a random variable that follows a normal distribution, where ξ y,i,t ~N(μ y,i,t ,σ y,i,t ). This paper uses the energy storage device of SESS-MEA to deal with the response uncertainty based on the chance constraint method. The method can be expressed as formula (16):

[0088]

[0089] Where: and are the estimated and actual values ​​of the total response of N microgrids respectively; The power of the energy storage device used to handle the response uncertainty; d i is the credit threshold of the i-th energy.

[0090] Furthermore, SESS-MEA constructs the first objective function based on the required response of different energy sources and considering the electric energy interaction with EVA;

[0091] In one embodiment, a stepped carbon trading system is used to divide CO2 emissions into several intervals to achieve effective restrictions on carbon emissions. The carbon emission sources of this model include power purchase from the power grid, CHP, and GB, among which power purchase from the power grid is an indirect emission source, and CHP and GB are direct carbon emission sources. In order to further improve the carbon emission reduction effect of the carbon trading mechanism and ensure the rationality of the interval selection of the mechanism, this application uses carbon quotas to describe the interval length and proposes a two-stage stepped carbon trading mechanism. This mechanism first uses the traditional stepped carbon trading mechanism to reduce the model carbon emissions and obtain the corresponding carbon emissions, and then refines the interval where the obtained carbon emissions are located and uses the stepped carbon emission mechanism after the refined interval to further reduce carbon emissions.

[0092]

[0093] E r =Ea -E e -E CHP -E GB =E a -E p (20)

[0094]

[0095] Where: C carbon is the carbon trading cost; e Carbon emission quota per unit of electricity for coal-fired units; is the electric power purchased by SESS-MEA from the grid; g,e With g,h Carbon emission quotas per unit of electricity and per unit of heat for gas-fired units; and are the output powers of GT, WHB and GB respectively; E e 、E CHP With E GB are the carbon emission quotas of coal-fired units, CHP and GB respectively; E p is the total carbon emission quota of SESS-MEA; E a With E r is the actual carbon emissions and carbon emissions trading volume; λ is the carbon trading base price; a is the initial interval range; b is the interval unit growth range; u is the transaction price growth range.

[0096] In this embodiment, the objective function of SESS-MEA is shown in formula (23):

[0097]

[0098] Where: C MEA is the total benefit of SESS-MEA; C b Profits from SESS-MEA's participation in demand response; to purchase costs for energy; C is the equipment operation and maintenance cost; re The cost of handling response uncertainty for energy storage devices; C f With C fa are the insufficient response penalty costs of microgrid and SESS-MEA, respectively.

[0099] 1) SESS-MEA demand response benefits

[0100]

[0101] Where: It is the response transaction price submitted by SESS-MEA to the trading center. is the response of SESS-MEA.

[0102] 2) Penalty cost for insufficient microgrid response

[0103]

[0104] Where: f i,t Penalty price for insufficient microgrid response.

[0105] 3) Energy purchase cost

[0106]

[0107] Where: and are the prices of RES purchased by SESS-MEA and electricity sold by the grid, respectively; and are natural gas and RES purchased power respectively.

[0108] 4) Equipment operation and maintenance costs

[0109]

[0110] Where: is the unit operation and maintenance cost of the energy conversion device in SESS-MEA; Output power to the energy conversion device; is the unit operation and maintenance cost of the energy storage device; Transmit power to energy storage devices.

[0111] 5) Dealing with the cost of response uncertainty

[0112]

[0113] Where: Unit costs of handling uncertainty for energy storage devices.

[0114] 6) SESS-MEA insufficient response penalty cost

[0115]

[0116] Embodiment 2

[0117] This embodiment establishes a demand response transaction model that considers the interaction between EVA and electric vehicles, and proposes the discomfort cost of electric vehicle demand response related to the differentiated incentive rate and response amount.

[0118] Specifically, the EV user determines the required response amount of electric energy based on the second differentiated incentive rate provided by the EVA; the EVA constructs a second objective function based on the required response amount of electric energy and taking into account the electric energy interaction with the SESS-MEA;

[0119] In the DR transaction model of EVA and EVs, EVA not only aggregates the response of EVs, but also provides charging services for them. In addition, to improve the applicability of the model, the model considers the impact of demand response volume and differentiated incentive rates on the discomfort cost of EVs user response.

[0120] First, for demand response DR, the response load of EVs is divided into IOL and TL, and the definition is consistent with the response load of the microgrid. The charging load of EVs after response is:

[0121]

[0122] Where: P EV,n,t and are the charging loads of the nth EV before and after the response at time t; is the k-th response load power of EV.

[0123] Furthermore, for the differentiated incentive rate provided by EVA, EVs determine the amount of electricity purchased from EVA and the response amount with the goal of minimizing the total cost. The objective function of EVs is shown in formula (34):

[0124]

[0125] Where: C n,EV is the total cost of the nth EV; c EVA,t and are the price and power of electricity purchased by EV from EVA at time t respectively; and They are the power and reward for EVs to actively participate in reducing electricity consumption during peak hours, which is provided by EVA; is the incentive rate of the κth response provided by EVA to EV; is the discomfort cost of EV participating in demand response, which is proportional to the response amount and inversely proportional to the incentive rate. Since TL only causes the transfer of EV load, it has a small impact on EV, while IOL has a greater impact on EV. Therefore, in order to describe the impact of the two responses on the discomfort cost of EV, the impact weight coefficient ω is introduced κ ;c dr Discomfort cost per unit.

[0126] In this embodiment, the power balance constraint of the EVs model is shown in formula (36):

[0127]

[0128] Secondly, EVA determines its energy purchase and sales power with the goal of maximizing revenue. The objective function is shown in formula (37):

[0129]

[0130] Where: C EVA is the total benefit of EVA; and The power of electricity purchased by EVA from the grid and the power of RES purchased from RES producers respectively; The price to buy RES for EVA.

[0131] Embodiment 3

[0132] Based on the objective functions disclosed in Example 1 and Example 2, this embodiment considers the interaction between SESS-MEA and EVA and establishes a multi-agent DR transaction game model. The DR transaction process based on this model is as follows: Figure 1 As shown, the decision variable of the electric energy interaction between the two is added; at this time, the objective functions (23) and (37) are transformed as follows:

[0133]

[0134]

[0135] Where: P ME,t With P EM,t are the electric power transmitted from SESS-MEA to EVA and the electric power transmitted from EVA to SESS-MEA at time t; according to the energy consumption characteristics of SESS-MEA and EVA, the price c of electricity sold by SESS-MEA to EVA is set ME,t It is lower during 21:00-6:00, and during 11:00-14:00, the price at which EVA sells electricity to SESS-MEA is set c EM,t Lower.

[0136] Finally, the objective functions of EVA and SESS-MEA are assigned weights q1 and q2, and the objective function C of the multi-agent game model is obtained. z :

[0137] C z =q1C EVA +q2C MEA (40)

[0138] In a preferred embodiment, since there is an obvious master-slave confrontation relationship between the microgrid and SESS-MEA, and between EVs and EVA, and SESS-MEA and EVA are interconnected, in order to improve the flexibility of the model and the penetration rate of renewable energy, a Steinberg game model is established to describe the response transaction process between the above subjects.

[0139] First, SESS-MEA and EVA, as leaders, determine the incentive rate and penalty price given to the microgrid and the incentive rate of EVs. Then, the microgrid and EVs determine the response amount respectively with the goal of minimizing the cost. Then, after receiving the response amount, SESS-MEA and EVA adjust the incentive rate and penalty price with the goal of maximizing the total benefit, and give it to the microgrid and EVs again. The process is repeated until the Nash solution of the model is obtained or the maximum number of iterations is reached. Since the model has many variables and the solution is complex, the particle swarm algorithm with high efficiency and fast convergence speed is used to solve it. The solution process is as follows Figure 2 shown.

[0140] In another embodiment, the step of determining the estimated response amount according to the target adjustment incentive rate and penalty price includes:

[0141] (1) SESS-MEA and MEA submit the demand response volume and incentive rate of microgrids and EVs to the trading market respectively;

[0142] (2) The trading market matches the information submitted by the responding demand party and the demand responding party, and publishes the clearing results.

[0143] In an exemplary embodiment, the DR transaction process can be generally divided into two stages, namely, response estimation and response bidding;

[0144] Since the DR transaction mechanism between EVA and EV users is consistent with the IDR transaction mechanism between SESS-MEA and microgrid, the IDR transaction between SESS-MEA and microgrid is taken as an example to introduce the response transaction process.

[0145] In the response estimation stage, SESS-MEA first provides differentiated incentive rates and penalty prices to each microgrid. Then, each microgrid determines the response amount of each energy source with the goal of minimizing cost, and submits the response amount to SESS-MEA. After receiving the response amount of the microgrid, SESS-MEA determines its response amount with the goal of maximizing profit. Finally, SESS-MEA submits the final response amount, transaction incentive rate, and transaction penalty price to the trading market. Among them, the transaction incentive rate and penalty price are the average of the incentive rate and penalty price of each microgrid. At the same time, the response demand direction submits the required response amount, transaction incentive rate, and penalty price to the trading market.

[0146] In the response bidding stage, the trading market matches SESS-MEA with the information submitted by the responding demand party and publishes the clearing result. The clearing result includes the response transaction volume, transaction incentive rate and transaction penalty price. Finally, the two successfully matched parties sign a contract and execute it the next day.

[0147] To verify the effect of the response method of this application, the present invention uses the Monte Carlo sampling method to obtain the charging load of 100 EVs, and sets the EV charging and discharging efficiency to 0.9. In the carbon trading mechanism, υe, υg,e and υg,h are set to 0.728, 0.392, 0.263 kg / (kW.h), λ=0.25 yuan / kg, u=20%, a=5%, b=2%. The price τ=0.4 for the microgrid to purchase electricity from SESS-MEA, and the response uncertainty ξy,i,t~N(1,0.67).

[0148] The comparative analysis of model superiority is as follows:

[0149] In order to verify the superiority of the responsive transaction model considering the interaction between SESS-MEA and EVA proposed in this paper, the following four scenarios are set for comparative analysis:

[0150] Scenario 1: There is no power interaction between MEA, microgrid, and EVA;

[0151] Scenario 2: There is power interaction between MEA and microgrid (SESS-MEA), but there is no power interaction between SESS-MEA and EVA;

[0152] Scenario 3: There is no power interaction between MEA and microgrid, but there is power interaction between MEA and EVA;

[0153] Scenario 4: There is power interaction between MEA and microgrid, and there is power interaction between SESS-MEA and EVA.

[0154] The operation results of the four scenarios are shown in Table 1. The RES consumption of the four scenarios is shown in Figure 3 As shown in the figure, the energy interaction between SESS-MEA and EVA in scenario 4 is as follows: Figure 4 shown.

[0155] Table 1

[0156] Scenario MEA income / yuan EVA income / yuan Microgrid cost / yuan Total cost / yuan Scenario 1 12089 48199 91582 31294 Scenario 2 11672 48199 90581 30710 Scene 3 14461 50332 91014 26221 Scene 4 13850 50152 89648 25646

[0157] As shown in Table 1, the MEA revenue and microgrid cost in Scenario 2 are both less than those in Scenario 1. This is because the microgrid in Scenario 2 improves its energy efficiency through interaction with MEA, reducing the energy purchase cost and energy storage cost of the microgrid in this scenario. In addition, since the MEA in Scenario 2 bears the power storage cost of each microgrid in the process of interacting with each microgrid and sells electricity to the microgrid at a lower price, the energy storage cost of MEA in Scenario 2 is higher than that in Scenario 1. However, since the reduction in microgrid cost in Scenario 2 is greater than the increase in MEA revenue, and the RES consumption in Scenario 2 is greater than that in Scenario 1, Scenario 2 is superior to Scenario 1 in improving RES penetration and increasing the total revenue of the model.

[0158] In scenario 3, MEA sells the remaining RES to EVA during the peak period of RES power generation, which increases the RES penetration rate and MEA's revenue, and reduces EVA's electricity purchase cost. EVA sells electricity to MEA at a lower price during the peak period of electricity consumption, which reduces MEA's energy purchase cost and increases MEA's revenue. Therefore, as shown in Table 1 and Figure 2 As shown in Figure 3, the benefits of MEA and EVA in scenario 3 are greater than those in scenario 1, and the RES penetration rate is higher than that in scenario 1. The above analysis shows that increasing the interaction between MEA and EVA is beneficial to improving the RES penetration rate and the economic efficiency of the model.

[0159] Depend on Figure 3 It can be seen that when considering the power interaction between MEA, microgrid and EVA at the same time, SESS-MEA provides power to EVA from 17:00 to 5:00, reducing the energy waste caused by SESS-MEA's difficulty in consuming all RES. EVA sells electricity to SESS-MEA during the low-load period of EV (7:00-16:00), reducing the power purchase cost of SESS-MEA and achieving efficient energy utilization. Figure 2 It can be obtained that, since there is electric energy interaction between MEA, microgrid and EVA in scenario 4, the MEA benefit of scenario 4 is greater than that of scenarios 1 and 2 but less than that of scenario 3, the microgrid cost and total cost are the smallest and the RES consumption is the highest.

[0160] In summary, increasing the interaction between MEA and microgrid and between MEA and EVA is beneficial to improving energy utilization and the benefits of each subject. Increasing the interaction between MEA and the other two subjects at the same time can further improve the economic efficiency of the model and the penetration rate of RES.

[0161] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0162] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-agent game demand response method considering the interaction between SESS-MEA and EVA, characterized in that: The microgrid determines the required response amounts of different energy sources based on the first differentiated incentive rate provided by the SESS-MEA; the SESS-MEA constructs a first objective function based on the penalty price and the required response amounts of the different energy sources and considering the electric energy interaction with the EVA; The microgrid determines the required response of different energy sources with the goal of minimizing cost; the formula is expressed as: Where: The cost of purchasing natural gas for the microgrid; c g,t is the price of natural gas; and They are the operation and maintenance costs of the energy conversion device and the energy storage device respectively; and are the unit operation and maintenance costs of the energy conversion device and the energy storage device, respectively; is the incentive rate given by SESS-ME to the κth load of the ith energy in the yth microgrid at time t; is the total transmission power of the energy storage device; is the output power of the ζ-th energy conversion device; Power purchased from natural gas for microgrids; The cost of purchasing electricity from SESS-MEA for the microgrid; Provide electrical power to SESS-MEA for the microgrid; The power given to the microgrid by SESS-MEA; is the response estimate of the κth type of response load of the ith energy load of the yth microgrid at time t; The first objective function is expressed as: Where: C MEA is the total benefit of SESS-MEA; C b Profits from SESS-MEA's participation in demand response; to purchase costs for energy; C is the equipment operation and maintenance cost; carbon is the carbon trading cost; C re The cost of handling response uncertainty for energy storage devices; C f With C fa are the insufficient response penalty costs of microgrid and SESS-MEA respectively; C ME is the revenue from electricity sales from SESS-MEA to EVA, C EM Costs incurred for purchasing electricity from EVA for SESS-MEA; The EV user determines the required response amount of electric energy based on the second differentiated incentive rate provided by the EVA; the EVA constructs a second objective function based on the required response amount of electric energy and taking into account the electric energy interaction with the SESS-MEA; The EV user determines the required response amount of electric energy with the goal of minimizing cost; the formula is expressed as: Where: is the total cost of the nth EV; c EVA，t and are the price and power of electricity purchased by EV from EVA at time t respectively; and They are the power and reward for EVs to actively participate in reducing electricity consumption during peak hours, which is provided by EVA; is the incentive rate of the κth response provided by EVA to EV; is the discomfort cost of EV participating in demand response, which is proportional to the response amount and inversely proportional to the incentive rate; is the k-th response load power of EV; is the charging load before the nth EV responds at time t; ω κ is the weight coefficient; c dr is the unit discomfort cost; The second objective function is expressed as: Where: C EVA is the total benefit of EVA; and The power of electricity purchased by EVA from the grid and the power of RES purchased from RES producers respectively; The price of purchasing RES for EVA; C ME is the revenue from electricity sales from SESS-MEA to EVA, C EM Costs incurred for purchasing electricity from EVA for SESS-MEA; is the price of electricity sold to the grid; EVA is the response transaction price submitted by EVA to the trading center; T is the scheduling period, and M is the total number of EVs; Weights are assigned to the first objective function and the second objective function to obtain a multi-agent game model, and the multi-agent game model is solved to obtain demand responses of the microgrid and EVs.

2. The multi-agent game demand response method according to claim 1, characterized in that: If the electric energy provided by the microgrid to SESS-MEA during the dispatch period is greater than the electric energy provided by SESS-MEA to the microgrid, SESS-MEA bears the cost of storing electric energy in the microgrid without paying the microgrid for that part of the electric energy; otherwise, the microgrid needs to pay SESS-MEA for the excess electric energy.

3. The multi-agent game demand response method according to claim 1, characterized in that: The carbon quota is used to describe the interval length and the carbon trading cost is determined through a two-stage tiered carbon trading mechanism. The public expression is: In the formula, E r is the carbon emission trading volume; λ is the carbon trading base price; u is the transaction price growth rate; E P is the total carbon emission quota of SESS-MEA; a is the initial interval range; b is the unit growth range of the interval.

4. The multi-agent game demand response method according to claim 1, characterized in that: Both the microgrid and SESS-MEA include energy conversion devices and energy storage devices. The energy conversion devices include: electric boilers, electric refrigerators, gas turbines, waste heat boilers, gas boilers, and absorption refrigerators.

5. The multi-agent game demand response method according to claim 1, characterized in that: Where: P ME,t With P EM,t are the electric power transmitted from SESS-MEA to EVA and the electric power transmitted from EVA to SESS-MEA at time t, respectively; c ME,t is the price of electricity sold by SESS-MEA to EVA; c EM,t The price at which EVA sells electricity to SESS-MEA.

6. The multi-agent game demand response method according to claim 1, characterized in that: Responding to the multi-agent demands based on the multi-agent game model includes: Solve the multi-agent game model to obtain the first target adjustment incentive rate, the target penalty price, and the second target adjustment incentive rate, and return them to the microgrid and the EVA user for repeated iterations until the iteration stop condition is met; And adjust the incentive rate and penalty price according to the iterative target to determine the estimated response amount.

Images