Optimal economic dispatching method for multi-agent integrated energy system

Abstract

An optimal economic dispatching method for a multi-subject integrated energy system comprises the following steps: step 1, constructing a double-layer master-slave game model based on an integrated demand response model and a multi-subject integrated energy system energy supply and income model; 2, generating a multi-agent comprehensive energy system, and reporting an energy selling unit price and a response compensation unit price to a user; the user performs comprehensive demand response according to the real-time electricity price and the incentive policy to obtain the actual electricity load and the energy purchase cost; 3, optimizing the output of the multi-main-body integrated energy system and each coupling device to obtain the minimum operation cost; 4, the profit of the multi-agent integrated energy system is solved through the double-layer master-slave game model constructed in the step 1, the upper layer of the double-layer master-slave game model adopts an adaptive particle swarm algorithm, and the lower layer of the double-layer master-slave game model adopts a mixed integer programming algorithm; and 5, iteration is carried out according to the process, the response compensation unit price is adjusted in real time until the total profit does not change any more, and an optimal economic dispatching scheme is output. According to the method, the optimal economy of the operation of the comprehensive energy system under the interaction of multiple interest subjects is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-agent integrated energy system optimization scheduling, and in particular to an optimal economic scheduling method for a multi-agent integrated energy system. Background Art

[0002] With the continuous increase in energy demand and the increasing severity of environmental pollution, safety, efficiency, low carbon and clean energy have become the mainstream direction of energy development. When centralized energy storage power stations and wind farms are built in the integrated energy and participate in the system dispatch as stakeholders, a multi-stakeholder integrated energy system is formed. In the multi-stakeholder integrated energy system transaction, system operators, energy supply operators and users are included. There are complex interactive relationships between the three, such as the two-way power interaction between the distribution network and each power consumption park, the one-way power interaction between the wind farm and the user, the two-way power interaction between the energy storage power station and each park, etc., and if the autonomous response behavior on the user side is considered, this brings difficulties to the economic optimization dispatch of the integrated energy system.

[0003] At present, some scholars have conducted research on the optimization and dispatching technology of integrated energy systems. The Chinese invention patent with application number 202010870015.9 proposed a multi-park integrated energy system optimization dispatching method based on master-slave game, which increased the amount of power interaction between microgrids and significantly reduced the electricity purchase cost of microgrids. The Chinese invention patent with application number 202110054775.7 proposed an optimization method for cluster trading of integrated energy systems at the park level, which maximized the revenue of energy-selling parks and minimized the cost of energy-purchasing parks.

[0004] Although many scholars have conducted research on the optimization and dispatching technology of integrated energy systems, many current studies have ignored the autonomous response behavior of the energy consumption side, or only studied a single type of demand response behavior, without fully considering the multi-load and multi-type integrated demand response behavior in the multi-agent integrated energy system. In addition, the existing master-slave game model for the integrated energy system fails to optimize the dispatch of centralized wind farms and energy storage power stations as interest subjects, and does not consider the impact of the alliance relationship between multiple interest subjects within the integrated energy system on the optimization results, resulting in the current limitations on the scope of application of the master-slave game model for the integrated energy system. When faced with an integrated energy system with a large number of distributed energy sources, and there are interactions between distributed energy sources or with the power grid, the number of variables involved in the optimization dispatch within the integrated energy system increases. If a variable is missing during the optimization dispatch process, the output result will not be the optimal solution. Summary of the invention

[0005] In view of the technical problems that the existing technologies fail to fully consider the load characteristics of power consumption parks and the alliance relationship between multiple stakeholders within the integrated energy system, resulting in the variables that affect the optimization results failing to participate in the optimization process of the system, which may lead to the user's energy purchase cost and the overall profit of the multi-agent integrated energy system failing to reach the optimal solution, this technical solution provides an optimal economic scheduling method for a multi-agent integrated energy system, and proposes a multi-category and multi-type comprehensive demand response with a more complete incentive surface according to the electricity consumption characteristics on the user side, including price-based electricity demand response for residential users, incentive-based electricity demand response for industrial users, fuzzy comfort cooling demand response for refrigerators, incentive-based heat demand response for heat-producing boilers, etc.; due to the multiple types of incentives With the addition of the policy, users can adjust their electricity load independently according to the subsidy to minimize the cost of energy purchase. Therefore, the positive interaction between the user side and the power grid is strengthened, unnecessary energy waste is reduced, and energy utilization is improved. Energy storage power stations and centralized wind farms are also involved in the optimal scheduling of multi-agent integrated energy systems, and energy storage power stations and centralized wind farm energy supply and revenue models are established. The alliance relationship within the multi-stakeholder integrated energy system and the supporting and complementary role of energy storage power stations to users are considered to address the problem of mismatch between new energy output and user-side load demand, thereby achieving the purpose of effective utilization of new energy grid-connected installed capacity, so that energy buyers and energy sellers can obtain satisfactory costs and profits, and improve the overall economy of the multi-stakeholder integrated energy system; effectively solving the above problems.

[0006] The present invention is achieved through the following technical solutions:

[0007] An optimal economic dispatch method for a multi-agent integrated energy system comprises the following steps:

[0008] Step 1: Based on the comprehensive demand response model and the multi-agent comprehensive energy system energy supply and benefit model, a two-layer master-slave game model is constructed;

[0009] The architecture of the multi-agent integrated energy system includes: a distribution network, an energy storage power station, a wind farm, and multiple power consumption parks, and there is an interaction of interests between the various entities; the multiple power consumption parks include photovoltaic power supply parks, residential user parks for combined heat and power, and industrial power consumption parks for combined heat and power;

[0010] Step 2: Generate a multi-agent integrated energy system and report the energy sales unit price and response compensation unit price to users; users respond to comprehensive demand based on real-time electricity prices and incentive policies to obtain actual electricity load and energy purchase costs;

[0011] Step 3: Optimize the output of the multi-agent integrated energy system and each coupled device to obtain the minimum operating cost;

[0012] Step 4: Using the two-layer master-slave game model constructed in step 1 to solve the profit of the multi-agent integrated energy system, the upper layer of the two-layer master-slave game model adopts an adaptive particle swarm algorithm, and the lower layer adopts a mixed integer programming algorithm;

[0013] Step 5: Iterate according to the above process and adjust the response compensation unit price in real time until the total profit no longer changes, and output the optimal economic scheduling plan.

[0014] Furthermore, the multiple power consumption parks described in step one adopt a comprehensive demand response method to increase the enthusiasm of the user side to interact with the power grid; the comprehensive demand response includes: price-based electricity demand response, incentive-based electricity demand response, cooling load demand response and incentive-based thermal load demand response; different users respond to demand based on the four demand response methods in the comprehensive demand response and their electricity consumption characteristics to reduce energy purchase costs.

[0015] Furthermore, the price-based electricity demand response corresponds to ordinary residential users in a residential user park with combined cooling, heating and power. The model expression of the price-based electricity demand response is:

[0016]

[0017] In the above formula, ε tt , ε th They represent the self-elasticity coefficient and the cross-elasticity coefficient, respectively, and are used to measure the relationship between the increase in electricity prices for users at different times; ρ t , Δρ t Respectively represent the electricity price and the change in electricity price at time t; ρ h , Δρ h They represent the electricity price and the change in electricity price at time h respectively; P t load,DR , P t 0 It represents the composite power of electricity after demand response and before demand response;

[0018] The incentive-type electricity demand response corresponds to the factory industrial users in the industrial power park of cogeneration. The model expression of the incentive-type electricity demand response is:

[0019]

[0020] In the above formula, C p,DR represents the incentive-based electricity demand response compensation income; T represents the total dispatch time; P t shift , They represent the transferable load at time t and the unit price of power transfer compensation at time t respectively; P t cut , Indicates the reducible electric load and the compensation price per unit electric power reduction at time t;

[0021] The cooling load demand response corresponds to the fuzzy comfort of refrigerators such as air conditioners and ice makers. The model expression of the cooling load demand response is:

[0022]

[0023] In the above formula, represents cooling load power; S represents cooling area; μ represents heat loss per unit temperature difference, and its value is 1.037×10 4 J; C is the heat capacity per unit cooling area, and its value is 1.63×10 5 J; W out,t , W in,t represents the outdoor and indoor temperature at time t; Δt represents the time change; W in,t-1 represents the indoor temperature at time t-1;

[0024] The incentive-type heat load demand response corresponds to heat-generating equipment such as air conditioning and heating and boilers. The model expression of the incentive-type heat load demand response is:

[0025]

[0026] In the above formula, C h,DR represents the thermal response compensation gain; They represent the load amount that can be transferred and the load amount that can be reduced at time t respectively; Indicates the unit price of heat load shift compensation and heat load reduction compensation.

[0027] Furthermore, the expression formula for the energy supply and revenue relationship of the energy storage power station in the multi-agent integrated energy system is:

[0028]

[0029]

[0030] In the above formula, C ess represents the unit power operation profit of the energy storage power station; T represents the total dispatch time; N represents the number of parks; t represents the time value from 1 to T, j represents the target park value from 1 to N; ψ represents the unit power operation cost of the energy storage power station; Indicates the comprehensive energy sales price of the energy storage power station; It represents the unit price of electric power sold per unit in park j at time t; P t ch ,wind It represents the power of energy storage power station discharging and charging to park j and charging from the power grid and wind farm; represents the basic energy selling price of the energy storage power station; τ represents the unit income of the auxiliary service of the energy storage power station;

[0031] Furthermore, the expression formula of the energy supply and benefit relationship of the wind farm in the multi-agent integrated energy system is:

[0032]

[0033] In the above formula, C wind , c wind,op They represent the unit power operating profit, electricity sales price and operating cost of the wind farm respectively; P t wind,grid , It represents the electricity sales power transmitted by the wind farm to the power grid and park j at time t; T represents the total dispatch time; N represents the number of parks; t represents the time value 1~T, and j represents the target park value 1-N;.

[0034] Furthermore, the internal power flow relationships of the multi-agent integrated energy system are as follows: there is a one-way power interaction between the distribution network and each of the multiple power consumption parks, there is a two-way power interaction between the energy storage power station and each of the multiple power consumption parks, there is a two-way power interaction between each of the multiple power consumption parks, and there is a one-way power interaction between the wind farm and the distribution network, the energy storage power station and each of the multiple power consumption parks.

[0035] Furthermore, the internal constraints of the multi-agent integrated energy system with power balance as the target are: (1) constraints on electric power balance, (2) constraints on thermal power balance, (3) constraints on cooling power balance, (4) constraints on power transmission between parks, (5) constraints on power transmission between parks and distribution networks, (6) constraints on the output of coupled equipment, (7) constraints on the operation of energy storage power stations, (8) constraints on the output of wind farms, and (9) constraints on the price and incentives for multi-agent energy to sell energy to users.

[0036] The expression of the electric power balance constraint is:

[0037]

[0038] in, They represent the total power after the response of park j and the power purchased by park j from the power grid; represents the power purchased and sold by park j to park i at time t; It represents the power consumption of distributed photovoltaic power generation and air conditioning in park j at time t; It represents the discharge power of the energy storage power station to park j at time t and the power generation power of the thermal power generation unit of park j at time t;

[0039] The expression of the constraint of the thermal power balance is:

[0040]

[0041] in, represents the total heat generation power after the response of park j; It is expressed as the heat production power of the cogeneration generator set and gas boiler in the park j at time t;

[0042] The expression of the constraint of the cold power balance is:

[0043]

[0044] in, It indicates the total cooling power of park j after the response; They represent the cooling power of the combined cooling, heating and power generating set and air conditioner in park j respectively;

[0045] The expression for the transmission power constraint between the parks is:

[0046]

[0047] in, It represents the power purchased and sold by the park j at time t; It represents the power purchased by park j from park i at time t, the upper limit of power purchased and the power purchasing behavior 0~1 variable; 0~1 variables representing the power sold by park j to park i, the upper limit of power sold and the power selling behavior;

[0048] The expression of the transmission power constraint between the park and the distribution network is:

[0049]

[0050] in, They represent the power purchased by park j from the distribution network at time t, the upper limit of power purchased, and the power purchasing behavior 0-1 variables; It represents the power sold by park j to the distribution network at time t, the upper limit of power sold, and the power selling behavior (0-1 variable);

[0051] The expression for the constraint of the coupling device output is:

[0052]

[0053] in, They represent the lower limit, upper limit, output power, and output behavior of device B in park j at time t, respectively, which are 0-1 variables; It represents the output power and output behavior of device B in park j at time t-1, which is a 0-1 variable; They represent the upper limit of the climbing power and the upper limit of the descending power of the device B in the park j respectively; They represent the start and stop behaviors of device B in park j at time t, 0 to 1 variables respectively;

[0054] The expression of the constraint of the energy storage power station operation is:

[0055]

[0056] in, Respectively represent the charge state of the energy storage power station at time t and time t-1; r ch 、r dis Respectively represent charging efficiency and discharging efficiency; P represents the power discharged by the energy storage power station to park j; t ch,grid Indicates the charging power of the energy storage station from the grid; P t ch,wind Indicates the charging power of the energy storage station from the wind farm; Indicates the charging power of the energy storage power station to park j; They represent the lower and upper limits of the state of charge of the energy storage power station respectively; They represent the charge state of the energy storage power station when t is equal to 1, T and the initial time respectively; They represent the maximum charging power and the charging behavior of the energy storage power station to park j at time t, which are 0-1 variables; They represent the maximum discharge power of the pure energy power station and the discharge behavior of the energy storage power station to park j at time t, which are 0-1 variables; They represent the power purchase behavior of the energy storage power station from the distribution network and wind farm at time t, respectively, which is a 0-1 variable;

[0057] The expression of the wind farm output constraint is:

[0058]

[0059] Among them, P t wind Indicates the total electricity sold by the wind farm; P t ch,wind , P t wind,grid They represent the power sold by the wind farm to park j at time t, the power charged by the energy storage power station from the wind farm at time t, and the power sold by the wind farm to the distribution network at time t; They represent the upper limit of the power sold by the wind farm to the park j and the power selling behavior of the wind farm to the park j at time t, which is a 0-1 variable; P wind,grid,max , They represent the upper limit of the power sold by the wind farm to the grid and the behavior variable of the wind farm selling electricity to the grid at time t, which is 0 to 1; P ch,wind,max , They represent the upper limit of the charging power of the energy storage power station to the wind farm and the charging behavior of the energy storage power station to the wind farm at time t, which are 0-1 variables;

[0060] The expression of the price and incentive constraints of the multi-agent energy selling energy to users is:

[0061]

[0062] in, They represent the unit price of electricity sold by the park to users at time t and the unit price of subsidies provided by the distribution network to users for transferring and reducing energy e; They represent the upper and lower limits of the corresponding unit price respectively.

[0063] Furthermore, the internal structure of the multi-agent integrated energy system aims at maximizing profits. The distribution network is set as the leader. The leader puts in a strategy set to release energy sales prices, response subsidies and equipment capacity in the park to users; the energy storage power station, wind farm and industrial power park are follower one, and the strategy set of follower one is charging and discharging power and energy supply power; the residential user park is follower two, and the strategy set of follower two is the reduction, transfer and replacement of various loads in response to comprehensive demand; the multi-agent integrated energy system pursues profit maximization, and its objective function is:

[0064]

[0065] Among them, F represents the total profit of the multi-agent integrated energy system; N represents the number of parks; T represents the total dispatch time; C j represents the profit of park j; C ess , C wind , C grid Represents the profits of energy storage power stations, wind farms and distribution networks; It represents the demand response of park users to energy e; e is the collection of electricity, heating and cooling loads; It represents the unit price of energy e sold by the park to users at time t; G represents the power purchased by park j from the power grid and the unit price of power sold by the power grid at time t; j,t It represents the gas purchasing power of park j at time t; represents the natural gas price at time t; represents the charging power of the energy storage station from park j; Indicates the unit price of electricity sold by the energy storage power station; It represents the power discharged by the energy storage power station to park j and the unit price of electricity purchased by the energy storage power station; It represents the power sold by the wind farm to park j and the unit price of the wind farm’s electricity; represents the unit power and unit power operating cost of device B in park j; C grid represents the profit of the distribution network; It represents the power and unit price of electricity sold by the distribution network to park j at time t; It represents the power and unit price of electricity sold by park j to the grid at time t; P represents the economic compensation paid by the distribution network to the users in park j for their demand response behavior; t ch,grid , P represents the charging power of the energy storage power station from the grid and the unit price of electricity purchased by the energy storage power station from the grid; t wind,grid , Indicates the power sold by the wind farm to the power grid and the unit price of electricity sold.

[0066] Furthermore, in the two-layer master-slave game model, the upper layer is the leader and the user, and the goal is to maximize the profit; the lower layer is the follower, and the goal is to minimize the cost; the upper layer uses an adaptive particle swarm algorithm to solve the user's energy purchase cost, and the lower layer uses a mixed integer programming algorithm to solve the cost of the multi-agent integrated energy system.

[0067] Furthermore, the residential user park and the industrial power park carry out diversified and multi-type comprehensive demand responses to the multi-subject integrated energy system according to the energy sales unit price and the response compensation unit price, and obtain the user's energy purchase cost and the income of the multi-subject integrated energy system; the lower layer of the two-layer master-slave game model optimizes the charging and discharging power of the energy storage power station and the output power of the coupling equipment according to the income of the multi-subject integrated energy system, and optimizes the power supply power of the wind farm at the same time, to obtain the comprehensive cost of the multi-subject integrated energy system, and finally calculates the daily total profit based on the income; iterate according to the process of steps one to four, if the next daily total profit is lower than the previous total profit, optimize the response subsidy unit price until the comprehensive profit of the multi-subject integrated energy system no longer changes; output the optimal economic scheduling plan.

[0068] Beneficial Effects

[0069] The optimal economic dispatch method of a multi-agent integrated energy system proposed in the present invention has the following beneficial effects compared with the prior art:

[0070] (1) The present invention establishes four types of comprehensive demand response models based on the electricity consumption characteristics of the user end, including a price-based electricity demand response model for residential users, an incentive-based electricity demand response model for industrial users, a fuzzy comfort cold demand response model for refrigerators, and an incentive-based heat demand response model for heat-generating equipment; different types of users can make appropriate demand responses according to their own electricity consumption characteristics, and conversely, the power grid can also accurately allocate electricity according to the user's demand response. At the same time, the centralized wind farms and energy storage power stations are taken as stakeholders to participate in the optimization and scheduling, and the complementary effects of energy storage power stations on park users are taken into consideration. The alliance relationship between multiple subjects within the integrated energy system is added to the master-slave game model for optimization, with the system operator as the leader, and the park load merchants, park users, wind farms and energy storage power stations as followers, seeking to maximize the overall profit of the multi-subject integrated energy system.

[0071] (2) The present invention expands the optimization scheduling of the integrated energy system including more subjects. Compared with the current research that only focuses on a single demand response behavior or ignores the user's autonomous response behavior, the multi-type integrated demand response established by the present invention improves the coverage of demand response. Different types of users can make appropriate demand responses according to their own electricity consumption characteristics. Conversely, the power grid can also accurately allocate electric energy according to the user's demand response. Finally, by iteratively optimizing the subsidy unit price, the enthusiasm of the user end and the power grid to interact is strengthened, so that energy is used more reasonably, energy waste is avoided, energy utilization and the overall economy of the integrated energy system of multiple stakeholders are improved, so that energy sellers can obtain satisfactory profits and energy buyers can obtain satisfactory costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 It is a schematic diagram of the overall process of the present invention.

[0073] Figure 2 This is a diagram of the optimized structure of the multi-stakeholder integrated energy system in the present invention.

[0074] Figure 3 This is a schematic diagram of the master-slave game interaction mechanism in the present invention.

[0075] Figure 4 This is a flow chart of economic optimization scheduling of the multi-stakeholder integrated energy system in the present invention. DETAILED DESCRIPTION

[0076] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Under the premise of not departing from the design concept of the present invention, various modifications and improvements made by ordinary persons in the art to the technical solutions of the present invention should all fall within the protection scope of the present invention.

[0077] Embodiment 1:

[0078] like Figure 1 As shown, a multi-agent integrated energy system optimal economic dispatch method includes the following steps:

[0079] Step 1: Based on the comprehensive demand response model and the multi-agent integrated energy system supply and benefit model, a two-layer master-slave game model is constructed.

[0080] like Figure 2 As shown in the figure, the architecture of the multi-agent integrated energy system includes: distribution network, energy storage power station, wind farm and multiple power consumption parks, and there is interest interaction between the various entities. The multiple power consumption parks include photovoltaic power supply parks, residential user parks with combined heat and power, and industrial power consumption parks with combined heat and power; Park 1 is for urban residential users including photovoltaic power generation, Park 2 is for industrial user parks with combined heat and power, and Park 3 is for urban residential user parks with combined heat and power. There is a two-way power interaction between parks and between parks and energy storage power stations, a one-way power interaction between the distribution network and the park, and a one-way power interaction between wind farms and other energy entities.

[0081] Taking into full consideration the characteristics of multiple power consumption parks, we adopt diversified and multi-type comprehensive demand response methods to improve the user side's enthusiasm for interaction with the power grid. Multiple power consumption parks adopt comprehensive demand response methods to improve the user side's enthusiasm for interaction with the power grid; comprehensive demand response includes: price-based electricity demand response, incentive-based electricity demand response, cooling load demand response and incentive-based thermal load demand response; different users respond to demand based on the four demand response methods in the comprehensive demand response according to their electricity consumption characteristics to reduce energy purchase costs.

[0082] The price-based electricity demand response corresponds to ordinary residential users in the residential user park of combined cooling, heating and power. The model expression of the price-based electricity demand response is:

[0083]

[0084] In the above formula, ε tt , ε th They represent the self-elasticity coefficient and the cross-elasticity coefficient, respectively, and are used to measure the relationship between the increase in electricity prices for users at different times; ρ t , Δρ t represents the electricity price and the change in electricity price at time t; ρ h , Δρ h represents the electricity price and the change in electricity price at time h; P t load,DR , P t 0 Represents the electric load power before and after demand response.

[0085] The incentive-type electricity demand response corresponds to the factory industrial users in the industrial power park of cogeneration. The model expression of the incentive-type electricity demand response is:

[0086]

[0087] In the above formula, C p,DR represents the incentive-based electricity demand response compensation income; T represents the total dispatch time; P t shift , P represents the transferable load and unit electric power transfer compensation price at time t; t cut , It indicates the reducible electric load and the compensation price per unit electric power reduction at time t.

[0088] The cooling load demand response corresponds to the fuzzy comfort of refrigerators such as air conditioners and ice makers. The model expression of the cooling load demand response is:

[0089]

[0090] In the above formula, represents cooling load power; S represents cooling area; μ represents heat loss per unit temperature difference, and its value is 1.037×10 4 J; C represents the heat capacity per unit cooling area, and its value is 1.63×10 5 J; W out,t , W in,t represents the outdoor and indoor temperature at time t; Δt represents the time change; W in,t-1 Represents the indoor temperature at time t-1.

[0091] The incentive-type heat load demand response corresponds to heat-generating equipment such as air conditioning and heating and boilers. The model expression of the incentive-type heat load demand response is:

[0092]

[0093] In the above formula, C h,DR represents the thermal response compensation gain; Indicates the amount of load that can be transferred and the amount of load that can be reduced at time t; They represent the unit price of heat load shift compensation and the unit price of heat load reduction compensation respectively.

[0094] The expression formula of the energy supply and benefit relationship of the energy storage power station in the multi-agent integrated energy system is:

[0095]

[0096]

[0097] In the above formula, C ess represents the unit power operation profit of the energy storage power station; T represents the total dispatch time; N represents the number of parks; t represents the time value from 1 to T, j represents the target park value from 1 to N; ψ represents the unit power operation cost of the energy storage power station; Indicates the comprehensive energy sales price of the energy storage power station; It represents the unit price of electric power sold per unit in park j at time t; P t ch ,wind It represents the power of energy storage power station discharging and charging to park j and charging from the power grid and wind farm; represents the basic energy selling price of the energy storage power station; τ represents the unit revenue of the auxiliary service of the energy storage power station.

[0098] The expression formula of the energy supply and benefit relationship of the wind farm in the multi-agent integrated energy system is:

[0099]

[0100] In the above formula, C wind , c wind,op They represent the unit power operating profit, electricity sales price and operating cost of the wind farm respectively; P t wind,grid , They represent the power sold by the wind farm to the power grid and park j at time t respectively; T represents the total dispatch time; N represents the number of parks; t represents the time value of 1 to T, and j represents the target park value of 1-N.

[0101] The internal power flow relationships of the multi-agent integrated energy system are as follows: there is a one-way power interaction between the distribution network and each of the multiple power consumption parks, there is a two-way power interaction between the energy storage power station and each of the multiple power consumption parks, there is a two-way power interaction between each of the multiple power consumption parks, and there is a one-way power interaction between the wind farm and the distribution network, the energy storage power station and each of the multiple power consumption parks.

[0102] Furthermore, the internal constraints of the multi-agent integrated energy system with power balance as the target are: (1) constraints on electric power balance, (2) constraints on thermal power balance, (3) constraints on cooling power balance, (4) constraints on power transmission between parks, (5) constraints on power transmission between parks and distribution networks, (6) constraints on the output of coupled equipment, (7) constraints on the operation of energy storage power stations, (8) constraints on the output of wind farms, and (9) constraints on the price and incentives for multi-agent energy to sell energy to users.

[0103] (1) The expression of the constraint of electric power balance is:

[0104]

[0105] in, It represents the total power after the response of park j and the power purchased by park j from the power grid; is the power purchased and sold by park j to park i at time t; It represents the power consumption of distributed photovoltaic power generation and air conditioning in park j at time t; It represents the discharge power of the energy storage power station to park j at time t and the power generation power of the thermal power generating set of park j at time t.

[0106] (2) The constraint expression of thermal power balance is:

[0107]

[0108] in, represents the total heat generation power after the response of park j; They represent the heat production power of the cogeneration generator set and gas boiler in the park j at time t respectively.

[0109] (3) The constraint expression of cooling power balance is:

[0110]

[0111] in, It indicates the total cooling power of park j after the response; It indicates the cooling power of the combined heating, cooling and power generating set and air conditioner in the park.

[0112] (4) The expression for the transmission power constraint between parks is:

[0113]

[0114]

[0115] in, It represents the power purchased and sold by the park j at time t; It represents the power purchased by park j from park i at time t, the upper limit of power purchased and the power purchasing behavior 0~1 variable; It is a 0-1 variable representing the power sold by park j to park i, the upper limit of power sold and the power selling behavior.

[0116] (5) The expression of the transmission power constraint between the park and the distribution network is:

[0117]

[0118] in, They represent the power purchased by park j from the distribution network at time t, the upper limit of power purchased, and the power purchasing behavior 0-1 variables; It is a 0-1 variable representing the power sold by park j to the distribution network at time t, the upper limit of the power sold, and the power selling behavior.

[0119] (6) The expression for the constraint of the coupling device output is:

[0120]

[0121] in, They represent the lower limit, upper limit, output power, and output behavior of device B in park j at time t, respectively, which are 0-1 variables; It represents the output power and output behavior of device B in park j at time t-1, which is a 0-1 variable; They represent the upper limit of the climbing power and the upper limit of the descending power of the device B in the park j respectively; They are 0~1 variables representing the start and stop behavior of device B in park j at time t.

[0122] (7) The expression of the constraints on the operation of the energy storage power station is:

[0123]

[0124] in, Respectively represent the charge state of the energy storage power station at time t and time t-1; r ch 、r dis Respectively represent charging efficiency and discharging efficiency; P represents the power discharged by the energy storage power station to park j; t ch,grid Indicates the charging power of the energy storage station from the grid; P t ch,wind Indicates the charging power of the energy storage station from the wind farm; Indicates the charging power of the energy storage power station to park j; They represent the lower and upper limits of the state of charge of the energy storage power station respectively; They represent the charge state of the energy storage power station when t is equal to 1, T and the initial time respectively; They represent the maximum charging power and the charging behavior of the energy storage power station to park j at time t, which are 0-1 variables; They represent the maximum discharge power of the pure energy power station and the discharge behavior of the energy storage power station to park j at time t, which are 0-1 variables; They represent the power purchase behavior of the energy storage power station from the distribution network and wind farm at time t, respectively, which are 0~1 variables.

[0125] (8) The expression of wind farm output constraint is:

[0126]

[0127] Among them, P twind represents the power sold by the wind farm at time t; P t ch,wind , P t wind,grid They represent the power sold by the wind farm to park j at time t, the power charged by the energy storage power station from the wind farm at time t, and the power sold by the wind farm to the distribution network at time t; are the upper limit of the power sold by the wind farm to the park j and the behavior variable of the wind farm selling power to the park j at time t, 0 to 1; P wind,grid,max , P is the upper limit of the power sold by the wind farm to the grid and the behavior variable of the wind farm selling electricity to the grid at time t, which is 0 to 1; ch,wind,max , It is a 0~1 variable that represents the upper limit of the charging power of the energy storage station to the wind farm and the charging behavior of the energy storage station to the wind farm at time t.

[0128] (9) The expression of the price and incentive constraints of energy sales from multiple energy entities to users is:

[0129]

[0130] in, They represent the unit price of electricity sold by the park to users at time t and the unit price of subsidies provided by the distribution network to users for transferring and reducing energy e; They represent the upper and lower limits of the corresponding unit price respectively.

[0131] The internal structure of the multi-agent integrated energy system aims to maximize profits. The distribution network is set as the leader. The leader's strategy set is to publish energy sales prices to users, respond to subsidies, and the capacity of equipment in the park; the energy storage power station, wind farm, and industrial power park are followers one, and the strategy set of follower one is charging and discharging power and energy supply power; the residential user park is follower two, and the strategy set of follower two is the reduction, transfer, and substitution of various loads in response to comprehensive demand; based on this, we build Figure 3 The master-slave game model of the multi-stakeholder integrated energy system shown in the figure is that the multi-stakeholder integrated energy system pursues profit maximization, and its objective function is:

[0132]

[0133] Among them, F represents the total profit of the multi-agent integrated energy system; N represents the number of parks; T represents the total dispatch time; C j represents the profit of park j; C ess , C wind , C grid Respectively represent the profits of energy storage power stations, wind farms and distribution networks; It represents the demand response of the users in the park to energy e; e represents the collection of electricity, heating and cooling loads; It represents the unit price of energy e sold by the park to users at time t; G represents the power purchased by park j from the power grid and the unit price of power sold by the power grid at time t; j,t It represents the gas purchasing power of park j at time t; represents the natural gas price at time t; represents the charging power of the energy storage station from park j; Indicates the unit price of electricity sold by the energy storage power station; It represents the power discharged by the energy storage power station to park j and the unit price of electricity purchased by the energy storage power station; It represents the power sold by the wind farm to park j and the unit price of the wind farm’s electricity; represents the unit power and unit power operating cost of device B in park j; C grid represents the profit of the distribution network; It represents the power and unit price of electricity sold by the distribution network to park j at time t; is the power sold by park j to the grid and the unit price of electricity sold at time t; P is the economic compensation paid by the distribution network to users in park j for their demand response behavior; t ch,grid , P represents the charging power of the energy storage power station from the grid and the unit price of electricity purchased by the energy storage power station from the grid; t wind,grid 、c t wind,grid Indicates the power sold by the wind farm to the power grid and the unit price of electricity sold.

[0134] Step 2: Generate a multi-agent integrated energy system and report the energy sales unit price and response compensation unit price to users; users make comprehensive demand responses based on real-time electricity prices and incentive policies to obtain actual electricity load and energy purchase costs.

[0135] Step 3: Optimize the output of the multi-agent integrated energy system and each coupled device to obtain the minimum operating cost.

[0136] Step 4: Use the two-layer master-slave game model constructed in step 1 to solve the profit of the multi-agent integrated energy system. The upper layer of the two-layer master-slave game model adopts an adaptive particle swarm algorithm, and the lower layer adopts a mixed integer programming algorithm.

[0137] In the two-layer master-slave game model, the upper layer is the leader and the user, and the goal is to maximize the profit; the lower layer is the follower, and the goal is to minimize the cost; the upper layer uses an adaptive particle swarm algorithm to solve the user's energy purchase cost, and the lower layer uses a mixed integer programming algorithm to solve the cost of the multi-agent integrated energy system.

[0138] Step 5: Iterate according to the above process and adjust the response compensation unit price in real time until the total profit no longer changes, and output the optimal economic scheduling plan.

[0139] Residential user parks and industrial power parks carry out diversified and multi-type comprehensive demand responses to the multi-agent integrated energy system according to the energy sales unit price and the response compensation unit price, and obtain the user's energy purchase cost and the income of the multi-agent integrated energy system; the lower layer of the two-layer master-slave game model optimizes the charging and discharging power of the energy storage power station and the output power of the coupling equipment according to the income of the multi-agent integrated energy system, and optimizes the power supply power of the wind farm at the same time, and obtains the comprehensive cost of the multi-agent integrated energy system, and finally calculates the daily total profit based on the income; iterate according to the process of steps one to four, if the next daily total profit is lower than the previous total profit, optimize the response subsidy unit price until the comprehensive profit of the multi-agent integrated energy system no longer changes; output the optimal economic scheduling plan.

[0140] According to Figure 4 The multi-stakeholder integrated energy system optimization scheduling plan process shown in FIG. is iterated, and the response subsidy unit price is optimized through iterative optimization until the comprehensive profit of the multi-stakeholder integrated energy system no longer changes, and the optimal economic scheduling plan is output.

[0141] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable people familiar with the technology to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any equivalent transformation or modification made according to the spirit of the present invention should be included in the present invention.

Claims

1. An optimal economic dispatch method for a multi-agent integrated energy system, characterized by: Includes steps: Step 1: Based on the comprehensive demand response model and the multi-agent comprehensive energy system energy supply and benefit model, a two-layer master-slave game model is constructed; The architecture of the multi-agent integrated energy system includes: a distribution network, an energy storage power station, a wind farm, and multiple power consumption parks, and there is an interaction of interests between the various entities; the multiple power consumption parks include photovoltaic power supply parks, residential user parks for combined heat and power, and industrial power consumption parks for combined heat and power; Step 2: Generate a multi-agent integrated energy system and report the energy sales unit price and response compensation unit price to users; users respond to comprehensive demand based on real-time electricity prices and incentive policies to obtain actual electricity load and energy purchase costs; Step 3: Optimize the output of the multi-agent integrated energy system and each coupled device to obtain the minimum operating cost; Step 4: Using the two-layer master-slave game model constructed in step 1 to solve the profit of the multi-agent integrated energy system, the upper layer of the two-layer master-slave game model adopts an adaptive particle swarm algorithm, and the lower layer adopts a mixed integer programming algorithm; Step 5: Iterate according to the above process and adjust the response compensation unit price in real time until the total profit no longer changes, and output the optimal economic scheduling plan.

2. The optimal economic dispatch method for a multi-agent integrated energy system according to claim 1 is characterized by: The multiple power consumption parks described in step 1 adopt a comprehensive demand response method to improve the enthusiasm of the user side to interact with the power grid; The comprehensive demand response includes: price-based electricity demand response, incentive-based electricity demand response, cooling load demand response and incentive-based heating load demand response; Based on the four demand response methods in the comprehensive demand response, different users respond to demand according to their electricity consumption characteristics to reduce energy purchase costs.

3. The optimal economic dispatch method for a multi-agent integrated energy system according to claim 2 is characterized by: The price-based electricity demand response corresponds to ordinary residential users in the residential user park of combined cooling, heating and power. The model expression of the price-based electricity demand response is: In the above formula, ε tt , ε th are the self-elasticity coefficient and the cross-elasticity coefficient, respectively, which are used to measure the relationship between the increase in electricity prices of users at different times; ρ t , Δρ t are the electricity price and the change in electricity price at time t respectively; ρ h , Δρ h are the electricity price and the change in electricity price at time h respectively; P t load,DR , P t 0 are the electric load power after and before demand response, respectively; The incentive-type electricity demand response corresponds to the factory industrial users in the industrial power park of cogeneration. The model expression of the incentive-type electricity demand response is: In the above formula, C p,DR represents the incentive-based electricity demand response compensation income; T represents the total dispatch time; P represents the transferable load and unit electric power transfer compensation price at time t; t cut , Indicates the reducible electric load and the compensation price per unit electric power reduction at time t; The cooling load demand response corresponds to the fuzzy comfort of refrigerators such as air conditioners and ice makers. The model expression of the cooling load demand response is: In the above formula, represents cooling load power; S represents cooling area; μ represents heat loss per unit temperature difference, and its value is 1.037×10 4 J; C represents the heat capacity per unit cooling area, and its value is 1.63×10 5 J; W out,t , W in,t represents the outdoor and indoor temperature at time t; Δt represents the time change; W in,t-1 represents the indoor temperature at time t-1; The incentive-type heat load demand response corresponds to heat-generating equipment such as air conditioning heating and boilers. The model expression of the incentive-type heat load demand response is: In the above formula, C h,DR represents the thermal response compensation gain; Indicates the amount of load that can be transferred and the amount of load that can be reduced at time t; Indicates the unit price of heat load shift compensation and heat load reduction compensation.

4. The optimal economic dispatch method for a multi-agent integrated energy system according to claim 1 is characterized by: The expression formula of the energy supply and benefit relationship of the energy storage power station in the multi-agent integrated energy system is: In the above formula, C ess represents the unit power operation profit of the energy storage power station; T represents the total dispatch time; N represents the number of parks; t represents the time value from 1 to T, j represents the target park value from 1 to N; ψ represents the unit power operation cost of the energy storage power station; Indicates the comprehensive energy sales price of the energy storage power station; It represents the unit price of electric power sold per unit in park j at time t; P t ch,grid , P t ch,wind It represents the power of energy storage power station discharging and charging to park j and charging from the power grid and wind farm; represents the basic energy selling price of the energy storage power station; τ represents the unit revenue of the auxiliary service of the energy storage power station.

5. The optimal economic dispatch method for a multi-agent integrated energy system according to claim 1 is characterized by: The expression formula of the energy supply and benefit relationship of the wind farm in the multi-agent integrated energy system is: In the above formula, C wind , c wind,op P represents the unit power operating profit, electricity sales price and operating cost of the wind farm; t wind,grid , It represents the power sold by the wind farm to the power grid and park j at time t; T represents the total dispatch time; N represents the number of parks; t represents the time value 1~T, and j represents the target park value 1-N.

6. The optimal economic dispatch method for a multi-agent integrated energy system according to claim 1 is characterized by: The internal power flow relationships of the multi-agent integrated energy system are as follows: there is a one-way power interaction between the distribution network and each of the multiple power consumption parks, there is a two-way power interaction between the energy storage power station and each of the multiple power consumption parks, there is a two-way power interaction between each of the multiple power consumption parks, and there is a one-way power interaction between the wind farm and the distribution network, the energy storage power station and each of the multiple power consumption parks.

7. The optimal economic dispatch method for a multi-agent integrated energy system according to claim 1 or 6, characterized in that: The internal constraints of the multi-agent integrated energy system with power balance as the target are: (1) constraints on electric power balance, (2) constraints on thermal power balance, (3) constraints on cooling power balance, (4) constraints on power transmission between parks, (5) constraints on power transmission between parks and distribution networks, (6) constraints on the output of coupled equipment, (7) constraints on the operation of energy storage power stations, (8) constraints on the output of wind farms, and (9) constraints on the price and incentives for multi-agent energy to sell energy to users. The expression of the electric power balance constraint is: in, It represents the total power of park j after response and the power purchased from the power grid; represents the power purchased and sold by park j to park i at time t; It represents the power consumption of distributed photovoltaic power generation and air conditioning in park j at time t; They represent the discharge power of the energy storage power station to park j at time t and the power generation power of the thermal power generation unit of park j at time t respectively; The expression of the constraint of the thermal power balance is: in, represents the total heat generation power after the response of park j; It represents the heat production power of the cogeneration generator set and gas boiler in the park j at time t; The expression of the constraint of the cold power balance is: in, It indicates the total cooling power of park j after the response; Indicates the cooling power of the combined cooling, heating and power generating set and air conditioner in the park; The expression for the transmission power constraint between the parks is: in, It represents the power purchased and sold by the park j at time t; It represents the power purchased by park j from park i at time t, the upper limit of power purchased and the power purchasing behavior 0~1 variable; 0~1 variables representing the power sold by park j to park i, the upper limit of power sold and the power selling behavior; The expression of the transmission power constraint between the park and the distribution network is: in, It represents the power purchased by park j from the distribution network at time t, the upper limit of power purchase, and the power purchase behavior (0-1 variable); It represents the power sold by park j to the distribution network at time t, the upper limit of power sold, and the power selling behavior (0-1 variable); The expression for the constraint of the coupling device output is: in, It represents the lower limit, upper limit, output power, and output behavior of device B in park j at time t, which are 0-1 variables; It represents the output power and output behavior of device B in park j at time t-1, which is a 0-1 variable; Indicates the upper limit of the climbing power and the upper limit of the descending power of the device B in the park j; A 0-1 variable representing the start and stop behavior of device B in park j at time t; The constraints of the energy storage power station operation and maintenance are: in, Respectively represent the charge state of the energy storage power station at time t and time t-1; r ch 、r dis Table charging efficiency and discharging efficiency; P represents the power discharged by the energy storage power station to park j; t ch,grid Indicates the charging power of the energy storage station from the grid; P t ch ,wind Indicates the charging power of the energy storage station from the wind farm; Indicates the charging power of the energy storage power station to park j; They represent the lower and upper limits of the state of charge of the energy storage power station respectively; Indicates the charge state of the energy storage power station when t equals 1, T and the initial time; They represent the maximum charging power and the charging behavior of the energy storage power station to park j at time t, which are 0-1 variables; They represent the maximum discharge power of the pure energy power station and the discharge behavior of the energy storage power station to park j at time t, which are 0-1 variables; They represent the power purchase behavior of the energy storage power station from the distribution network and wind farm at time t, respectively, which is a 0-1 variable; The expression of the wind farm output constraint is: Among them, P t wind represents the power sold by the wind farm at time t; P t ch,wind , P t wind,grid They represent the power sold by the wind farm to park j at time t, the power charged by the energy storage power station from the wind farm at time t, and the power sold by the wind farm to the distribution network at time t; They represent the upper limit of the power sold by the wind farm to the park j and the behavior variable of the wind farm selling power to the park j at time t, which is 0 to 1; P t wind,grid,max , They represent the upper limit of the power sold by the wind farm to the grid and the behavior variable of the wind farm selling electricity to the grid at time t, which is 0 to 1; P ch,wind,max , They represent the upper limit of the charging power of the energy storage power station to the wind farm and the charging behavior of the energy storage power station to the wind farm at time t, which are 0-1 variables; The expression of the constraints of the energy selling price and incentives of the multi-agent integrated energy system to users is: in, They represent the unit price of electricity sold by the park to users at time t and the unit price of subsidies provided by the distribution network to users for transferring and reducing energy e; They represent the upper and lower limits of the corresponding unit price respectively.

8. The optimal economic dispatch method for a multi-agent integrated energy system according to claim 1 is characterized by: The internal structure of the multi-agent integrated energy system aims at maximizing profits. The distribution network is set as the leader. The leader’s strategy set is to release the energy selling price, response subsidy and equipment capacity in the park to users; the energy storage power station, wind farm and industrial power park are follower one, and the strategy set of follower one is charging and discharging power and energy supply power; the residential user park is follower two, and the strategy set of follower two is the reduction, transfer and replacement of various loads in response to comprehensive demand; the multi-agent integrated energy system pursues profit maximization, and its objective function is: Among them, F represents the total profit of the multi-agent integrated energy system; N represents the number of parks; T represents the total dispatch time; C j represents the profit of park j; C ess , C wind , C grid Represents the profits of energy storage power stations, wind farms and distribution networks; It represents the demand response of park users to energy e; e is the collection of electricity, heating and cooling loads; It represents the unit price of energy e sold by the park to users at time t; G represents the power purchased by park j from the power grid and the unit price of power sold by the power grid at time t; j,t It represents the gas purchasing power of park j at time t; represents the natural gas price at time t; represents the charging power of the energy storage station from park j; Indicates the unit price of electricity sold by the energy storage power station; It represents the power discharged by the energy storage power station to park j and the unit price of electricity purchased by the energy storage power station; It represents the power sold by the wind farm to park j and the unit price of the wind farm’s electricity; represents the unit power and unit power operating cost of device B in park j; C grid represents the profit of the distribution network; It represents the power and unit price of electricity sold by the distribution network to park j at time t; It represents the power and unit price of electricity sold by park j to the grid at time t; P represents the economic compensation paid by the distribution network to the users in park j for their demand response behavior; t ch,grid , P represents the charging power of the energy storage power station from the grid and the unit price of electricity purchased by the energy storage power station from the grid; t wind,grid , Indicates the power sold by the wind farm to the power grid and the unit price of electricity sold.

9. The optimal economic dispatch method for a multi-agent integrated energy system according to claim 1, characterized in that: In the two-layer master-slave game model, the upper layer is the leader and the user, and the goal is to maximize the profit; the lower layer is the follower, and the goal is to minimize the cost; the upper layer uses an adaptive particle swarm algorithm to solve the user's energy purchase cost, and the lower layer uses a mixed integer programming algorithm to solve the cost of the multi-agent integrated energy system.

10. The optimal economic dispatch method for a multi-agent integrated energy system according to claim 9, characterized in that: The residential user park and the industrial power park perform diversified and multi-type comprehensive demand responses to the multi-subject integrated energy system according to the energy sales unit price and the response compensation unit price, and obtain the user energy purchase cost and the income of the multi-subject integrated energy system; The lower layer of the two-layer master-slave game model optimizes the charging and discharging power of the energy storage power station and the output power of the coupling equipment according to the benefits of the multi-agent integrated energy system, and optimizes the power supply power of the wind farm at the same time, so as to obtain the comprehensive cost of the multi-agent integrated energy system, and finally calculates the daily total profit in combination with the benefits; iterates according to the process of steps one to four, and if the next daily total profit is lower than the previous total profit, the response subsidy unit price is optimized until the comprehensive profit of the multi-agent integrated energy system no longer changes; Output the optimal economic dispatch plan.

Images