Multi-microgrid multi-level demand response method considering fraudulent behavior

Abstract

The invention belongs to the technical field of micro-grid energy demand response, and particularly relates to a multi-micro-grid multi-level demand response method considering fraudulent behaviors, and the method comprises the steps: building a multi-micro-grid multi-level distributed architecture; establishing a micro-grid energy system model; establishing a micro-grid user low-carbon economic optimization operation model; establishing a micro-grid energy operator optimization clearing model and a user low-carbon economic optimization model considering the IDR; establishing a non-cooperative game model in the microgrid IDR market and solving the non-cooperative game model; establishing and solving a load aggregator alliance peak regulation demand distribution strategy optimization model; solving a transaction mode involved in the micro-grid user low-carbon economic optimization operation model; and solving the multi-level distributed demand response optimization model to obtain a multi-microgrid multi-level demand response strategy considering the fraudulent behavior. According to the method, the fraud phenomenon between users in the micro-grid is fully considered, the benefit subjects are coupled, and the economical efficiency of the whole alliance is optimal.

Description

Technical Field

[0001] The present invention belongs to the technical field of microgrid energy demand response, and in particular relates to a multi-microgrid multi-level demand response method taking fraudulent behavior into consideration. Background Art

[0002] The development of energy structure presents a diversified trend. Consumers need to integrate different energy sources, improve energy efficiency, and make full use of existing controllable resources. In this context, the integrated energy system (IES) came into being. While IES is booming, the demand side gradually shows characteristics such as large load demand, complex load characteristics, and high requirements for power system reliability. Traditional demand response can no longer meet the system requirements and gradually develops towards integrated demand response (IDR). The operation of the future power grid will largely rely on the integration of multi-energy networks. IDR combines load reduction, translation and energy type conversion methods, which can help the demand side achieve energy supply and demand balance in a more flexible way and deeply stimulate the response potential of the demand side.

[0003] At present, there have been many research results on IDR at home and abroad, and the research perspectives mainly include time scale, market mechanism, and control strategy. Since IDR integrates multiple energy sources, the time scale of IDR is relatively complex, mainly divided into short-term and medium- and long-term. Short-term IDR is of great significance for system scheduling, market transactions, and optimized operation of distributed networks within the next 48 hours: for example, in existing studies, considering IDR and fuel cell-based hydrogen storage systems, short-term robust optimization scheduling of hydrogen-based intelligent micro-energy centers is proposed. Medium- and long-term IDR includes monthly planning and annual planning, which are greatly affected by seasons and policies: for example, in existing studies, by combining the power generation planning model with the unit combination economic scheduling model, the long-term comprehensive impact of renewable energy and demand-side flexible and adjustable resources on thermal power generation operations is studied.

[0004] In existing research, a low-carbon economic operation optimization model for energy hubs has also been proposed. Under the premise of considering equipment losses, long-term dynamic planning is implemented for energy hubs that integrate IDR, distributed photovoltaics and energy storage devices. In response to the problem of fraud in the game process, microgrids will deliberately provide false information for their own interests, but this may lead to the breakdown of cooperative operation of multiple microgrids and affect the overall benefits of the multi-microgrid system. Therefore, it is necessary to establish a fraud balance mechanism between microgrids and use it to respond to multi-microgrid multi-level demand. Summary of the invention

[0005] In view of the above deficiencies in the prior art, the purpose of the present invention is to provide a multi-microgrid multi-level demand response method that takes fraudulent behavior into consideration, deeply explore the demand response capability of the power system through three levels: multi-microgrid alliance, microgrid as a whole, and microgrid users, fully consider the fraud phenomenon between users in the microgrid, couple various stakeholders, and achieve the optimal economic efficiency of the entire alliance.

[0006] To achieve the above objectives, the present invention provides a multi-microgrid multi-level demand response method considering fraudulent behavior, comprising the following steps: S1. Establish a multi-microgrid multi-level distributed architecture. The multi-microgrid multi-level distributed architecture includes two levels, namely, a load aggregator alliance based on cooperative game and multiple microgrid IDR markets based on non-cooperative game (which has the characteristics of decentralized self-optimization). The load aggregator alliance includes multiple load aggregators, and each microgrid IDR market includes two levels, namely, a microgrid energy operator and multiple users; S2. Based on the microgrid IDR market, a microgrid energy system model is established. The microgrid energy system model includes four energy forms: electricity, gas (mainly natural gas), heat, and cold. The user equipment involved is divided into energy production equipment, energy conversion equipment, and energy storage equipment, and a mathematical model of each device is established; S3. With the goal of minimizing daily operating costs, establish a low-carbon economic optimization operation model for microgrid users and set its constraints; S4. Based on the microgrid IDR market, establish an optimal clearing model for microgrid energy operators and a low-carbon economic optimization model for users considering IDR; S5. Establish a non-cooperative game model in the microgrid IDR market based on the microgrid energy operator's optimal clearing model and the user's low-carbon economic optimization model considering IDR, and solve the non-cooperative game model; S6. Establish a load aggregator alliance peak load demand allocation strategy optimization model and solve it; S7. Solve the transaction mode (mainly fraud cost and fraud factor) involved in the low-carbon economic optimization operation model of microgrid users; S8, multi-microgrid multi-level distributed architecture, microgrid energy system model, microgrid user low-carbon economic optimization operation model, non-cooperative game model and load aggregator alliance peak demand allocation strategy optimization model together constitute a multi-level distributed demand response optimization model. The multi-level distributed demand response optimization model is solved to obtain a multi-microgrid multi-level demand response strategy that takes fraud into account.

[0007] As a preferred solution of the present invention, in S1, users in the microgrid IDR market should not only consider the price game with other users that affects their probability of winning the bid, but also maximize the economic benefits brought by their participation in the microgrid IDR market, achieve low-carbon economic optimization in a non-cooperative game, and report the optimized energy consumption plan to the microgrid energy operator; The microgrid energy operator plays a leading role, aggregating users' energy consumption plans and issuing response quantities to each user based on the microgrid peak-shaving demand released by the load aggregator; the microgrid energy operator collects the quotations and declared peak-shaving ratios of each user, determines the winning capacity of each user and the clearing price of the microgrid IDR market, clears the microgrid IDR market with the goal of minimizing the overall IDR cost of the microgrid, and distributes the profits after the settlement is successful.

[0008] As a preferred embodiment of the present invention, in S2, the energy generation equipment includes a combined heat and power system, a gas boiler and an air conditioner, wherein the combined heat and power system includes a gas turbine and a waste heat boiler, and its mathematical model is expressed as follows: (1); In the formula, The electrical power output for the combined heat and power system; The medium-grade thermal power output of the combined heat and power system; The low-grade thermal power output for the combined heat and power system; The natural gas consumption for the combined heat and power system; is the power generation efficiency of the gas turbine; , They are the medium-quality thermal efficiency and low-quality thermal efficiency of the waste heat boiler respectively; is the calorific value of natural gas; T is time (usually 24, representing 24 hours); Gas boilers heat water by burning natural gas to produce high-temperature steam. Its mathematical model is expressed as: (2); In the formula, is the output thermal power of the gas boiler; is the natural gas consumption of gas boilers; is the heat generation efficiency of the gas boiler; The mathematical model of air conditioning is expressed as: (3); In the formula, is the cooling power of the electric refrigerator; is the power consumption of the electric refrigerator; is the energy efficiency ratio of the electric refrigerator; The energy conversion equipment includes a steam-driven device, an absorption refrigerator, and an AC / DC converter. The steam-driven device is driven by medium-grade hot steam and low-grade hot steam is recovered. Its mathematical model is expressed as follows: (4); In the formula, Low-grade thermal power recovered for steam-driven equipment; is the steam load; Heat recovery efficiency for steam-driven equipment; The mathematical model of the absorption chiller is expressed as: (5); In the formula, is the cooling power of the absorption chiller; is the heat consumption power of the absorption refrigerator; is the energy efficiency ratio of the absorption chiller; The mathematical model of the AC-DC converter is expressed as: (6); In the formula, and are the AC side input power and DC side output power of the AC / DC converter respectively; and are the conversion efficiencies from AC to DC and from DC to AC, respectively; The energy storage device includes an electric storage device and an ice storage device, wherein the mathematical model of the electric storage device is expressed as: (7); In the formula, and are the battery power at time t and time t+1 respectively; and are the charging power and discharging power of the battery at time t respectively; is the self-consumption coefficient of the battery; and are the charging efficiency and discharging efficiency of the battery respectively; The mathematical model of the ice storage device is expressed as: (8); In the formula, , and are the cooling powers of the ice storage device, refrigerator and ice storage tank respectively; , and are the power consumption of the ice storage device, the refrigerator and the ice storage tank respectively; is the energy efficiency ratio of the refrigerator; and are the ice storage capacity of the ice storage tank at time t and at time t+1 respectively; is the self-consumption coefficient of the ice storage tank; and They are the ice making energy efficiency ratio and refrigeration efficiency of the ice storage tank respectively; is the power consumption of the ice storage tank at time t; is the cooling power of the ice storage tank at time t.

[0009] As a preferred solution of the present invention, in S3, the objective function of the low-carbon economic optimization operation model of the microgrid user is: (9); In the formula, is the daily operating cost of user i; is the energy cost of user i; Equipment operation and maintenance costs for user i; The cost of starting and stopping the equipment of user i; is the interruption load cost of user i; is the carbon trading cost of user i; is the fraud cost of user i; T is time; in, The calculation formula is: (10); In the formula, is the electricity price at time t; is the price of natural gas; is the price of hot steam; is the power purchased by user i at time t; is the gas purchasing power of user i at time t; is the purchased heat power of user i at time t; The calculation formula is: (11); In the formula, The operation and maintenance cost per unit output power of the j-th device of user i; is the output power of the jth device of user i at time t; The calculation formula is: (12); In the formula, The startup and shutdown costs of the jth device of user i; and are the status of the j-th device of user i at time t and time t-1 respectively. The status value is 1 for power on and 0 for shutdown; The calculation formula is: (13); In the formula, is the interruption load of user i at time t; For peak load time; and is the cost coefficient corresponding to the interruptible load; The calculation formula is: (14); In the formula, is the actual carbon emissions of user i; is the initial carbon emission quota of user i; is the unit carbon emission trading price; The calculation formula is: (15); In the formula, For fraud factors; is the cost difference before and after user i participates in the response interaction.

[0010] As a preferred solution of the present invention, in S3, the constraint conditions of the low-carbon economic optimization operation model of microgrid users are: Power balance constraints: The operation of the user's integrated energy system must meet the power balance constraints of electricity, heat, cooling and gas, specifically: (16); (17); (18); (19); (20); In the formula, , , , are the AC load, DC load, heating load and cooling load of user i at time t respectively; Produce power for photovoltaics; is the output power of the jth gas turbine of user i at time t; is the power consumption of the jth ice storage tank of user i at time t; is the AC output power of the jth device of user i at time t; is the AC conversion power of the jth device of user i at time t; is the DC conversion power of the jth device of user i at time t; is the battery charging power of the j-th device of user i at time t; is the battery discharge power of the jth device of user i at time t; is the thermal power of the output of the j-th device of user i at time t; is the output low-quality thermal power of the j-th device of user i at time t; is the output thermal power of the jth device of user i at time t; is the low-grade thermal power recovered by the j-th device of user i at time t; is the steam load of the j-th device of user i at time t; is the heat consumption power of the jth device of user i at time t; is the cooling power of the electric refrigerator of the j-th device of user i at time t; is the cooling power of the ice storage device of the jth device of user i at time t; is the cooling power of the absorption chiller of the jth device of user i at time t; Equipment operation constraints, where the equipment itself has power upper and lower limit constraints and ramp rate constraints: (twenty one); (twenty two); In the formula, , are the upper and lower limits of the output power of the j-th device of user i, respectively; is the maximum climbing rate of the j-th device of user i; is the output power of the jth device of user i at time t-1; The energy storage capacity constraint of the energy storage device is: (twenty three); (twenty four); In the formula, and They are the highest and lowest power of the battery respectively; and They are the upper and lower limits of ice storage capacity of the ice storage tank respectively; is the battery capacity of user i at the jth time t; is the ice storage capacity of the j-th ice storage tank of user i at time t; and are the starting and ending power of the j-th battery of user i respectively; and are the starting ice storage capacity and ending ice storage capacity of the j-th ice storage tank of user i respectively; The constraints for energy storage devices to be charged and discharged at different times are: (25); (26); (27); (28); In the formula, is the maximum power of the battery; and is the introduced 0-1 variable; and are the maximum electrical power and maximum refrigeration power of the ice storage tank respectively; The fraud factor constraint is expressed as: (29); In the formula, is the fraud factor limit; N is the number of users in the microgrid; It is the sum of all microgrid operating costs; It is the average cost of operating all microgrids.

[0011] Dishonest microgrids all hope that their fraud factors can reach the upper limit to obtain the optimal benefit. The upper limit of the fraud factor and the optimal benefit will change dynamically due to the influence of their own and other microgrid fraud behaviors. Therefore, each microgrid should strive to achieve fraud equilibrium. When it is very close to 0, each user reaches a fraud equilibrium and there is an optimal fraud factor Under such a fraud equilibrium, no user in the microgrid can increase the fraud factor alone to obtain higher profits without affecting the interests of other microgrids. At this time, all microgrids believe that the fraud choices they make can maximize their own profits.

[0012] As a preferred solution of the present invention, in the S4, in the microgrid IDR market, the information interaction between the microgrid energy operator and each user is used to trade the response quantity as a commodity within the microgrid, so as to achieve the purpose of minimizing the overall cost of the microgrid on the basis of user self-optimization; after receiving the response quantity allocated by the microgrid energy operator, the user re-optimizes his own energy consumption plan considering IDR, and reports the optimized IDR quotation and response quantity to the microgrid energy operator; the microgrid energy operator is responsible for the operation and settlement of the microgrid IDR market, and determines the response quantity allocation plan based on the quotation information and response capacity of each user, with the goal of optimal economic performance, and sends each user's winning bid quantity and IDR market clearing price to each user; Specifically, the optimization clearing model of the microgrid energy operator is that when the power grid issues a peak-shaving demand, the microgrid energy operator allocates the peak-shaving amount that users should respond to in each time period, which can be expressed as: (30); In the formula, is the response amount of the i-th user at time t; is the planned power purchase of the i-th user at time t without considering IDR; is the maximum available capacity of the i-th user at time t, which is calculated as follows: (31); In the formula, is the peak load demand of the power grid at time t; is the maximum demand of the ith user; In the microgrid IDR market, the microgrid energy operator reasonably allocates the response volume at each peak-shaving moment according to the peak-shaving ratio and quotation reported by each user, determines the user's winning capacity and the IDR market clearing price. The decision-making goal of the microgrid energy operator is to minimize the total IDR cost of all users in the microgrid while ensuring that the total response volume of the microgrid meets the peak-shaving demand of the power grid. The objective function is: (32); In the formula, is the total IDR cost of the microgrid supplier; is the set of supplier numbers, Quote the IDR for the i-th user; is the peak load ratio of the i-th user at time t; The peak load regulation amount of all users who win the bid should be equal to the peak load regulation demand of the power grid, that is, it meets the following requirements: (33); The microgrid energy operator issues the clearing price at the maximum of all bidders’ bids. Users do not need to bear the peak load regulation of the power grid, so the power grid should compensate for this part of the response. After the microgrid IDR market settlement is completed, the microgrid energy operator is responsible for distributing the income obtained from the power grid to each user, which is expressed as: (34); In the formula, The grid compensation fee allocated to the i-th user; The grid compensation fee obtained for the entire microgrid is calculated as follows: (35); In the formula, Charges for peak load regulation services for the power grid; is the planned power purchase of the i-th user at time t after considering IDR; The user low-carbon economic optimization model considering IDR is that the user responds to the peak load demand of the power grid, and its operation optimization should consider the benefits in the IDR market. The objective function is: (36); In the formula, The total cost after considering IDR for the i-th user; is the income obtained by the i-th user in the microgrid IDR market, and the calculation method is as follows: (37); In the formula, It is the clearing price for the microgrid IDR market; In order to prevent users from exhausting the adjustable resources at a certain moment and to ensure that they have responsiveness at all peak-shaving moments, add constraints: (38); In the formula, is a constant greater than 0, which means that the peak load ratio of user i at all peak load times is the same; In the microgrid IDR market, there are two types of users: suppliers and demanders. The actual IDR response of suppliers is greater than the maximum available capacity, and the excess can be sold to gain revenue. The actual IDR response of demanders is less than the maximum available capacity, and they need to purchase the insufficient IDR capacity. If it is greater than 1, it means that user i is a supplier in the microgrid IDR market. is positive, otherwise it is the demand side. is negative; After each user reports the response volume to the microgrid energy operator, the supplier also needs to declare its expected IDR price. , the IDR quotation of each supplier is declared according to the average cost of the IDR capacity actually sold, and the calculation method is as follows: (39); In the formula, The peak load ratio When it is 1, the total cost of the i-th user after considering IDR; is the total cost of the i-th user after considering IDR.

[0013] As a preferred solution of the present invention, in S5, the non-cooperative game model in the microgrid IDR market is that in the microgrid IDR market, the final winning bid of each user is determined by the bid amount and the bid price declared by all the users participating in the bidding. There is a game on the bid price among the users, and the game result determines the probability of winning the bid in the IDR market in the park and then determines its interests. That is, the interests of the users are affected by the decision-making behavior of other participants. Therefore, there is a non-cooperative game relationship between the users in the microgrid IDR market. The game model can be characterized as ; N b For game participants, that is, all users in the microgrid IDR market, ; S is the strategy available to all participants in the microgrid IDR market in the game. ; is the daily electricity purchase plan of user i, U is the utility function of all participants, ,in, is the utility function of user i, which is calculated as follows: (40); In the non-cooperative game model in the microgrid IDR market, the existence of non-cooperative game equilibrium is explained as follows: The utility function of each participant in the microgrid IDR market is not only related to the strategy he chooses, but also depends on the strategies of other participants. Assuming that under a certain strategy combination, no participant can act alone to increase revenue, this strategy combination is called Nash equilibrium; In a multi-party non-cooperative game problem, if the strategy set of each participant is a non-empty compact convex subset of the Euclidean space, and the utility function is continuous quasi-concave in its strategy space, then the game problem has a Nash equilibrium; The game strategy in the non-cooperative game of the microgrid IDR market is the daily electricity purchase plan of each user, which needs to meet the constraints of the microgrid user low-carbon economic optimization operation model and the user low-carbon economic optimization model considering IDR. When user i participates in the microgrid IDR market bidding, there must be a corresponding daily electricity purchase plan, so the non-cooperative game strategy set is a non-empty compact convex subset; The utility function of user i is shown in formula (40): , If both are convex functions with respect to energy use strategies, then the linear superposition utility function of the two parts is a continuous quasi-concave function under the corresponding strategies; The solution process of the non-cooperative game model is: S51. The microgrid energy operator initially sets the peak-shaving service price of the power grid as the settlement price of the microgrid IDR market; the user reports the optimal peak-shaving ratio and IDR quotation under the current clearing price based on the user low-carbon economic optimization model considering IDR; S52, the microgrid energy operator summarizes the total peak load reported by all users and determines whether it is not greater than the peak load demand issued by the power grid, that is, whether formula (41) is established. If so, the IDR market is settled according to the peak load service price of the power grid, and the peak load ratio of each user is implemented according to its reported value, and the solution is completed; if not, continue to step 3, where formula (41) is: (41); S53. The microgrid energy operator determines the IDR market clearing price and the total cost of fraud within the microgrid based on the microgrid energy operator's optimized clearing model; S54. The user reports the optimal peak load ratio and IDR quotation under the current clearing price based on the user low-carbon economic optimization model considering IDR; S55. The microgrid energy operator determines whether the formula is valid. If not, it returns to S53. If yes, the IDR market is settled according to the clearing price of the previous iteration, and the user's winning capacity is such that the total IDR amount = peak-shaving demand.

[0014] As a preferred solution of the present invention, in S6, the load aggregator alliance peak-shaving demand allocation strategy optimization model is: the load aggregator is referred to as an aggregator, and each aggregator forms an alliance, with the goal of minimizing the total cost of all microgrids managed by the alliance, to determine the optimal peak-shaving demand allocation strategy, and the objective function is: (42); In the formula, The total cost of the microgrid managed by the aggregator alliance after participating in IDR; S uni A collection of alliances formed by aggregators; is the total cost of the microgrid managed by aggregator m after participating in IDR; a microgrid managed by the alliance is a microgrid IDR market based on non-cooperative game; The optimization variable of the load aggregator alliance peak load demand allocation strategy optimization model is the peak load demand allocated to each microgrid (which can be expressed as ), the total cost of each microgrid after participating in IDR is expressed as: (43); The profit distribution mechanism is based on the distribution of power grid revenue, and the calculation formula is: (44); In the formula, The revenue obtained from the grid by the microgrid managed by aggregator m after participating in IDR; is the cumulative value of the maximum demand of all users in the microgrid managed by aggregator m; is the cumulative value of the purchased power of all users of the microgrid managed by aggregator m at time t before alliance optimization; is the cumulative value of the purchased power of all users of the microgrid managed by aggregator m at time t after alliance optimization; is the total amount of peak demand transferred within the alliance; is the increased revenue of multi-microgrids due to cooperative game, and the calculation formula is: (45); In the formula, is the total revenue of the alliance; is the profit of aggregator m before participating in the cooperative game; The response amount of users in the microgrid cannot be negative. Based on this premise, the upper and lower limits of the peak-shaving demand allocation are set, that is, the allocation of peak-shaving demand should meet the following requirements: (46); In the formula, is the peak load demand allocated to microgrid m at time t; The PSO algorithm is used to solve the load aggregator alliance peak demand allocation strategy optimization model.

[0015] As a preferred solution of the present invention, in S7, the process of solving the transaction mode is as follows: S71. Initialize fraud cost for each user , iteration number k=1, fraud factor =0.001, convergence accuracy =0.0001; S72. Each user calculates the fraud cost based on Fraud costs Upload to microgrid energy operators; S73. The microgrid energy application provider summarizes the fraud costs and feeds back the sum of the fraud costs to each user; S74, each user based on Update the fraud factor, where is the relaxation coefficient, and its value range is ; is the kth iteration ; is the fraud factor baseline value; S75, judge convergence, if it satisfies , output the total fraud cost, otherwise k=k+1, return to S72.

[0016] As a preferred solution of the present invention, in S8, the solution process of the multi-level distributed demand response optimization model is: S81. A coalition of load aggregators based on cooperative game constitutes an alliance optimization layer. Based on the alliance optimization layer, an initial particle group is randomly generated to distribute the peak demand allocation strategy to each microgrid. S82. Each microgrid completes the microgrid IDR market clearing based on non-cooperative game and the aggregation of the total fraud cost within the microgrid under the current peak-shaving demand allocation strategy, and returns the total cost of the microgrid to the load aggregator alliance; S83, the alliance optimization layer updates the historical optimal particle and the global optimal population according to the information returned by the microgrid, updates the position and speed of the particles, and generates a new particle swarm; S84. The alliance optimization layer sends the new particle swarm to the microgrid layer for cyclic iterative calculation. When the number of iterations reaches the specified value, the cycle stops and the last generation of population is output.

[0017] The model and algorithm involved in the present invention can be executed by an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the software to implement the above-mentioned model and algorithm calculation.

[0018] The beneficial effects of the present invention are: The present invention designs a demand response model and solution method, which mainly aims at the problems that the demand side gradually shows large load demand, complex load characteristics, high requirements for power system reliability and fraud in the game. With the goal of maximizing the demand response capability of the system, the multi-layer model is jointly solved, and the optimal strategy is used to guide microgrid users to perform optimal control.

[0019] The present invention proposes a cost accounting model that takes fraudulent behavior into consideration during the game process, and sequentially establishes decision-making models for load aggregators, microgrid energy operators, and microgrid users, including a microgrid IDR market bidding model of a non-cooperative game based on the existing two-part electricity price, an optimized clearing model of microgrid energy operators, and a multi-microgrid alliance peak-shaving demand allocation strategy optimization model based on a cooperative game. By jointly solving the multi-layer model, the optimal strategy is used to guide the optimized scheduling of microgrid users.

[0020] The present invention deeply explores the demand response capability of the system through three levels: multi-microgrid alliance, microgrid as a whole, and microgrid users, fully considers the fraud phenomenon between users in the microgrid, couples various stakeholders, achieves the optimal economy of the entire alliance, and makes all individuals profitable. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a schematic diagram of the process of the present invention; Figure 2is a schematic diagram of a multi-microgrid multi-level distributed architecture in an embodiment of the present invention; Figure 3 It is a schematic diagram of the overall architecture of a single microgrid IDR market in an embodiment of the present invention; Figure 4 It is a flow chart of demand response optimization in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The embodiments of the present invention are further described below in conjunction with the accompanying drawings: like Figure 1 As shown, a multi-microgrid multi-level demand response method considering fraudulent behavior includes the following steps: S1. Establish a multi-microgrid multi-level distributed architecture. The multi-microgrid multi-level distributed architecture includes two levels, namely, a load aggregator alliance based on cooperative game and multiple microgrid IDR markets based on non-cooperative game. The load aggregator alliance includes multiple load aggregators. Each microgrid IDR market includes two levels, namely, a microgrid energy operator and multiple users. S2. Based on the microgrid IDR market, a microgrid energy system model is established. The microgrid energy system model includes four energy forms: electricity, gas, heat, and cold. The user equipment involved is divided into energy production equipment, energy conversion equipment, and energy storage equipment (based on the actual functions and operating characteristics of the user equipment), and a mathematical model of each device is established; S3. With the goal of minimizing daily operating costs, establish a low-carbon economic optimization operation model for microgrid users and set its constraints; S4. Based on the microgrid IDR market, establish an optimal clearing model for microgrid energy operators and a low-carbon economic optimization model for users considering IDR; S5. Establish a non-cooperative game model in the microgrid IDR market based on the microgrid energy operator's optimal clearing model and the user's low-carbon economic optimization model considering IDR, and solve the non-cooperative game model; S6. Establish a load aggregator alliance peak load demand allocation strategy optimization model and solve it; S7. Solve the transaction mode involved in the low-carbon economic optimization operation model of microgrid users; S8, multi-microgrid multi-level distributed architecture, microgrid energy system model, microgrid user low-carbon economic optimization operation model, non-cooperative game model and load aggregator alliance peak demand allocation strategy optimization model together constitute a multi-level distributed demand response optimization model. The multi-level distributed demand response optimization model is solved to obtain a multi-microgrid multi-level demand response strategy that takes fraud into account.

[0023] The multi-microgrid multi-level distributed architecture in S1 is as follows Figure 2As shown, the microgrid represents a microgrid IDR market based on non-cooperative game.

[0024] The multi-microgrid multi-level distributed architecture involves three layers of stakeholders: load aggregators, microgrid energy operators, and microgrid users. There are different regulatory requirements and interest games between the various entities. There are multiple stakeholders of the same type who compete with each other within the two levels of load aggregators and users. Each user in the microgrid belongs to a different stakeholder. As an independent and rational entity, each microgrid hopes to maximize the response benefits through cooperation. This embodiment uses the Nash negotiation model to enable each participating microgrid in the multi-microgrid system to obtain the Pareto optimal benefit.

[0025] Among them, load aggregators are a new type of role in the power market, mainly responsible for centralized management and optimized distribution of customer-side power loads to achieve more efficient power operation and utilization. Load aggregators use technical means to aggregate scattered small power loads to form a larger total load, so that they can operate and manage more efficiently. They are an important form of power demand response in the power system, balancing supply and demand by adjusting power demand and improving the stability and efficiency of the power system.

[0026] In S1, users in the microgrid IDR market must not only consider the price game with other users that affects their bid probability, but also maximize their economic benefits from participating in the microgrid IDR market, achieve low-carbon economic optimization in a non-cooperative game, and report the optimized energy use plan to the microgrid energy operator; The microgrid energy operator plays a leading role, aggregating users' energy consumption plans and issuing response quantities to each user based on the microgrid peak-shaving demand released by the load aggregator; the microgrid energy operator collects the quotations and declared peak-shaving ratios of each user, determines the winning capacity of each user and the clearing price of the microgrid IDR market, clears the microgrid IDR market with the goal of minimizing the overall IDR cost of the microgrid, and distributes the profits after the settlement is successful.

[0027] In S2, the energy generation equipment includes a combined heat and power system, a gas boiler and an air conditioner, wherein the combined heat and power system includes a gas turbine and a waste heat boiler, the energy conversion equipment includes a steam drive device, an absorption chiller and an AC / DC converter, wherein the steam drive device is driven by medium-grade hot steam and low-grade hot steam is recovered, and the energy storage equipment includes an electric storage device and an ice storage device (the ice storage device usually includes two main parts: a chiller and an ice storage tank). The overall architecture reference diagram of a single microgrid IDR market is as follows: Figure 3 This architecture realizes the cascade utilization of thermal energy, fully utilizes the circulating waste heat, and improves energy utilization.

[0028] In S3, the objective function of the low-carbon economic optimization operation model of microgrid users is: (9); In the formula, is the daily operating cost of user i; is the energy cost of user i; Equipment operation and maintenance costs for user i; The cost of starting and stopping the equipment of user i; is the interruption load cost of user i; is the carbon trading cost of user i; is the fraud cost of user i; T is time; in, The calculation formula is: (10); In the formula, is the electricity price at time t; is the price of natural gas; is the price of hot steam; is the power purchased by user i at time t; is the gas purchasing power of user i at time t; is the purchased heat power of user i at time t; The calculation formula is: (11); In the formula, The operation and maintenance cost per unit output power of the j-th device of user i; is the output power of the jth device of user i at time t; The calculation formula is: (12); In the formula, The startup and shutdown costs of the jth device of user i; and are the status of the j-th device of user i at time t and time t-1 respectively. The status value is 1 for power on and 0 for shutdown; The calculation formula is: (13); In addition to using multi-energy complementary methods, users can also respond to time-of-use electricity prices by controlling interruptible loads, where: is the interruption load of user i at time t; For peak load time; and is the cost coefficient corresponding to the interruptible load, which is related to the user's production situation; The calculation formula is: (14); In the formula, is the actual carbon emissions of user i; is the initial carbon emission quota of user i; is the unit carbon emission trading price; The calculation formula is: (15); In the formula, For fraud factors; is the cost difference before and after user i participates in the response interaction.

[0029] In S3, the constraints of the low-carbon economic optimization operation model of microgrid users are set.

[0030] In S4, in the microgrid IDR market, the response volume is traded as a commodity within the microgrid by utilizing the information interaction between the microgrid energy operator and each user, so as to achieve the goal of minimizing the overall cost of the microgrid based on the self-optimization of the user; after receiving the response volume allocated by the microgrid energy operator, the user re-optimizes his own energy consumption plan considering the IDR, and reports the optimized IDR quotation and response volume to the microgrid energy operator; the microgrid energy operator is responsible for the operation and clearing of the microgrid IDR market, and determines the response volume allocation plan based on the quotation information and response capacity of each user, with the goal of optimal economic performance, and sends each user's winning bid and IDR market clearing price to each user; Specifically, the optimization clearing model of microgrid energy operators is that when the power grid issues peak-shaving demand, the microgrid energy operator allocates the peak-shaving amount that users should respond to in each period. At present, most industrial users are subject to a two-part electricity price, which includes a capacity price and an electricity price. The calculation basis of the capacity price is the capacity or maximum demand of the user. The microgrid energy operator can allocate its peak-shaving amount to respond to by referring to the maximum demand of each user, which is expressed as: (30); In the formula, is the response amount of the i-th user at time t; is the planned power purchase of the i-th user at time t without considering IDR; is the maximum available capacity of the i-th user at time t, which is calculated as follows: (31); In the formula, is the peak load demand of the power grid at time t; is the maximum demand of the ith user; In the microgrid IDR market, the microgrid energy operator reasonably allocates the response volume at each peak-shaving moment according to the peak-shaving ratio and quotation reported by each user, determines the user's winning capacity and the IDR market clearing price. The decision-making goal of the microgrid energy operator is to minimize the total IDR cost of all users in the microgrid while ensuring that the total response volume of the microgrid meets the peak-shaving demand of the power grid. The objective function is: (32); In the formula, is the total IDR cost of the microgrid supplier; is the set of supplier numbers, Quote the IDR for the i-th user; is the peak load ratio of the i-th user at time t; The peak load of all users who win the bid should be equal to the peak load demand of the power grid, that is, it meets the following requirements: (33); The microgrid energy operator issues the clearing price at the maximum of all bidders’ bids. Users do not need to bear the peak load regulation of the power grid, so the power grid should compensate for this part of the response. After the microgrid IDR market settlement is completed, the microgrid energy operator is responsible for distributing the income obtained from the power grid to each user, which is expressed as: (34); In the formula, The grid compensation fee allocated to the i-th user; The grid compensation fee obtained for the entire microgrid is calculated as follows: (35); In the formula, Charges for peak load regulation services for the power grid; is the planned power purchase of the i-th user at time t after considering IDR; The user low-carbon economic optimization model considering IDR is that the user responds to the peak load demand of the power grid, and its operation optimization should consider the benefits in the IDR market. The objective function is: (36); In the formula, The total cost after considering IDR for the i-th user; is the income obtained by the i-th user in the microgrid IDR market, and the calculation method is as follows: (37); In the formula, It is the clearing price for the microgrid IDR market; In order to prevent users from exhausting the adjustable resources at a certain moment and to ensure that they have responsiveness at all peak-shaving moments, add constraints: (38); In the formula, is a constant greater than 0, which means that the peak load ratio of user i at all peak load times is the same; In the microgrid IDR market, there are two types of users: suppliers and demanders. The actual IDR response of suppliers is greater than the maximum available capacity, and the excess can be sold to gain revenue. The actual IDR response of demanders is less than the maximum available capacity, and they need to purchase the insufficient IDR capacity. If it is greater than 1, it means that user i is a supplier in the microgrid IDR market. is positive, otherwise it is the demand side. is negative; After each user reports the response volume to the microgrid energy operator, the supplier also needs to declare its expected IDR price. , the IDR quotation of each supplier is declared according to the average cost of the IDR capacity actually sold, and the calculation method is as follows: (39); In the formula, The peak load ratio When it is 1, the total cost of the i-th user after considering IDR; is the total cost of the i-th user after considering IDR.

[0031] In S5, the non-cooperative game model in the microgrid IDR market is: in the microgrid IDR market, the final winning bid of each user is determined by the bid amount and the bid price submitted by all the users participating in the bidding. There is a game on the bid price among the users. The game result determines the probability of winning the bid in the IDR market in the park and then determines its interests. That is, the interests of users are affected by the decision-making behavior of other participants. Therefore, there is a non-cooperative game relationship between users in the microgrid IDR market. The game model can be characterized as ; N b For game participants, that is, all users in the microgrid IDR market, ; S is the strategy available to all participants in the microgrid IDR market in the game. ; is the daily electricity purchase plan of user i, U is the utility function of all participants, ,in, is the utility function of user i, which is calculated as follows: (40); In the non-cooperative game model in the microgrid IDR market, the existence of non-cooperative game equilibrium is explained as follows: The utility function of each participant in the microgrid IDR market is not only related to the strategy selected by himself, but also depends on the strategies of other participants. Assuming that under a certain strategy combination, no participant can act alone to increase revenue, this strategy combination is called Nash equilibrium. Whether Nash equilibrium exists or not requires verification of whether the established non-cooperative model meets certain specific conditions. In a multi-party non-cooperative game problem, if the strategy set of each participant is a non-empty compact convex subset of the Euclidean space, and the utility function is continuous quasi-concave in its strategy space, then the game problem has a Nash equilibrium; The game strategy in the non-cooperative game of the microgrid IDR market is the daily electricity purchase plan of each user, which needs to meet the constraints of the microgrid user low-carbon economic optimization operation model and the user low-carbon economic optimization model considering IDR. When user i participates in the microgrid IDR market bidding, there must be a corresponding daily electricity purchase plan, so the non-cooperative game strategy set is a non-empty compact convex subset; The utility function of user i is shown in formula (40): , If both are convex functions with respect to energy use strategies, then the linear superposition utility function of the two parts is a continuous quasi-concave function under the corresponding strategies; The solution process of the non-cooperative game model is as follows (it can be solved using CPLEX software): S51. The microgrid energy operator initially sets the peak-shaving service price of the power grid as the settlement price of the microgrid IDR market; the user reports the optimal peak-shaving ratio and IDR quotation under the current clearing price based on the user low-carbon economic optimization model considering IDR; S52, the microgrid energy operator summarizes the total peak load reported by all users and determines whether it is not greater than the peak load demand issued by the power grid, that is, whether formula (41) is established. If so, the IDR market is settled according to the peak load service price of the power grid, and the peak load ratio of each user is implemented according to its reported value, and the solution is completed; if not, continue to step 3, where formula (41) is: (41); S53. The microgrid energy operator determines the IDR market clearing price and the total cost of fraud within the microgrid based on the microgrid energy operator's optimized clearing model; S54. The user reports the optimal peak load ratio and IDR quotation under the current clearing price based on the user low-carbon economic optimization model considering IDR; S55. The microgrid energy operator determines whether the formula is valid. If not, it returns to S53. If yes, the IDR market is settled according to the clearing price of the previous iteration, and the user's winning capacity is such that the total IDR amount = peak-shaving demand.

[0032] The design of this market mechanism takes into account the existing two-part electricity price billing method and improves its practical feasibility. In addition, under this mechanism, the non-cooperative game relationship between users can prevent users from raising prices by retaining peak-shaving capacity, which is conducive to motivating users to fully tap their IDR potential.

[0033] In S6, the load aggregator alliance peak-shaving demand allocation strategy optimization model is: the load aggregator is referred to as the aggregator. Each aggregator forms an alliance, and the goal is to minimize the total cost of all microgrids managed by the alliance to determine the optimal peak-shaving demand allocation strategy. The objective function is: (42); In the formula, is the total cost of the microgrid managed by the aggregator alliance after participating in IDR; S uni A collection of alliances formed by aggregators; is the total cost of the microgrid managed by aggregator m after participating in IDR; a microgrid managed by the alliance is a microgrid IDR market based on non-cooperative game; The optimization variable of the load aggregator alliance peak-shaving demand allocation strategy optimization model is the peak-shaving demand allocated to each microgrid. The total cost of each microgrid after participating in IDR is expressed as: (43); The key point of non-cooperative game research is how participants can maximize their own interests in the process of playing games with other entities, that is, to choose the best strategy in a situation where interests influence each other; while the premise of cooperative game is that the overall benefits of the alliance after cooperation are greater than the overall benefits of all participants before cooperation, which is a kind of collective rationality. Under the premise of maximizing the interests of the alliance, the cooperative game must also make all participants profitable, otherwise the alliance will break down, so the focus of non-cooperative game research is how to distribute the interests of the alliance.

[0034] The profit distribution mechanism is based on the distribution of power grid revenue, and the calculation formula is: (44); In the formula, The revenue obtained from the grid by the microgrid managed by aggregator m after participating in IDR; is the cumulative value of the maximum demand of all users in the microgrid managed by aggregator m; is the cumulative value of the purchased power of all users of the microgrid managed by aggregator m at time t before alliance optimization; is the cumulative value of the purchased power of all users of the microgrid managed by aggregator m at time t after alliance optimization; is the total amount of peak demand transferred within the alliance; is the increased revenue of multi-microgrids due to cooperative game, and the calculation formula is: (45); In the formula, is the total revenue of the alliance; is the profit of aggregator m before participating in the cooperative game; The response amount of users in the microgrid cannot be negative. Based on this premise, the upper and lower limits of the peak-shaving demand allocation are set, that is, the allocation of peak-shaving demand should meet the following requirements: (46); In the formula, is the peak load demand allocated to microgrid m at time t; The PSO algorithm is used to solve the load aggregator alliance peak demand allocation strategy optimization model.

[0035] In order to prevent fraud between users from causing the breakdown of the microgrid system energy operator, this embodiment proposes a microgrid energy operator transaction model in which each microgrid participates in fraud negotiations, which can achieve fraud equilibrium and effectively avoid privacy leakage of each participating entity. The transaction model requires a third party that all users trust to summarize the fraud costs and feed back the results to each user in the microgrid.

[0036] In S7, the solution process of the transaction mode is: S71. Initialize fraud cost for each user , iteration number k=1, fraud factor =0.001, convergence accuracy =0.0001; S72. Each user calculates the fraud cost based on Fraud costs Upload to microgrid energy operators; S73. The microgrid energy application provider summarizes the fraud costs and feeds back the sum of the fraud costs to each user; S74, each user based on Update the fraud factor, where is the relaxation coefficient, and its value range is ; is the kth iteration ; is the fraud factor benchmark value; the relaxation coefficient value range can refer to the known technology; S75, judge convergence, if it satisfies , output the total fraud cost, otherwise k=k+1, return to S72.

[0037] In S8, this embodiment adopts a particle swarm algorithm to solve, and the solution process of the multi-level distributed demand response optimization model is: S81. A coalition of load aggregators based on cooperative game constitutes an alliance optimization layer. Based on the alliance optimization layer, an initial particle group is randomly generated to distribute the peak demand allocation strategy to each microgrid. S82. Each microgrid completes the microgrid IDR market clearing based on non-cooperative game and the aggregation of the total fraud cost within the microgrid under the current peak-shaving demand allocation strategy, and returns the total cost of the microgrid to the load aggregator alliance; S83, the alliance optimization layer updates the historical optimal particle and the global optimal population according to the information returned by the microgrid, updates the position and speed of the particles, and generates a new particle swarm; S84. The alliance optimization layer sends the new particle swarm to the microgrid layer for cyclic iterative calculation. When the number of iterations reaches the specified value, the cycle stops and the last generation of population is output.

[0038] In the above process, the overall optimization process of multi-level distributed demand response is as follows: Figure 4 shown.

Claims

1. A multi-microgrid multi-level demand response method considering fraudulent behavior, characterized in that The following steps are involved: S1. Establish a multi-microgrid multi-level distributed architecture. The multi-microgrid multi-level distributed architecture includes two levels, namely, a load aggregator alliance based on cooperative game and multiple microgrid IDR markets based on non-cooperative game. The load aggregator alliance includes multiple load aggregators. Each microgrid IDR market includes two levels, namely, a microgrid energy operator and multiple users. S2. Based on the microgrid IDR market, a microgrid energy system model is established. The microgrid energy system model includes four energy forms: electricity, gas, heat, and cold. The user equipment involved is divided into energy production equipment, energy conversion equipment, and energy storage equipment, and a mathematical model of each device is established; S3. With the goal of minimizing daily operating costs, establish a low-carbon economic optimization operation model for microgrid users and set its constraints; S4. Based on the microgrid IDR market, establish an optimal clearing model for microgrid energy operators and a low-carbon economic optimization model for users considering IDR; S5. Establish a non-cooperative game model in the microgrid IDR market based on the microgrid energy operator's optimal clearing model and the user's low-carbon economic optimization model considering IDR, and solve the non-cooperative game model; S6. Establish a load aggregator alliance peak load demand allocation strategy optimization model and solve it; S7. Solve the transaction mode involved in the low-carbon economic optimization operation model of microgrid users; S8, multi-microgrid multi-level distributed architecture, microgrid energy system model, microgrid user low-carbon economic optimization operation model, non-cooperative game model and load aggregator alliance peak demand allocation strategy optimization model together constitute a multi-level distributed demand response optimization model. The multi-level distributed demand response optimization model is solved to obtain a multi-microgrid multi-level demand response strategy that takes fraud into account.

2. A multi-microgrid multi-level demand response method considering fraudulent behavior according to claim 1, characterized in that: In S1, users in the microgrid IDR market must not only consider the price game with other users that affects their bid probability, but also maximize their economic benefits from participating in the microgrid IDR market, achieve low-carbon economic optimization in non-cooperative games, and report the optimized energy use plan to the microgrid energy operator; The microgrid energy operator plays a leading role, aggregating users' energy consumption plans and issuing response quantities to each user based on the microgrid peak-shaving demand released by the load aggregator; the microgrid energy operator collects the quotations and declared peak-shaving ratios of each user, determines the winning capacity of each user and the clearing price of the microgrid IDR market, clears the microgrid IDR market with the goal of minimizing the overall IDR cost of the microgrid, and distributes the profits after the settlement is successful.

3. A multi-microgrid multi-level demand response method considering fraudulent behavior according to claim 1, characterized in that: In S2, the energy generation equipment includes a combined heat and power system, a gas boiler and an air conditioner, wherein the combined heat and power system includes a gas turbine and a waste heat boiler, and its mathematical model is expressed as: （1）； In the formula, The electrical power output for the combined heat and power system; The medium-grade thermal power output of the combined heat and power system; The low-grade thermal power output for the combined heat and power system; The natural gas consumption for the combined heat and power system; is the power generation efficiency of the gas turbine; , They are the medium-quality thermal efficiency and low-quality thermal efficiency of the waste heat boiler respectively; is the calorific value of natural gas; T is time; Gas boilers heat water by burning natural gas to produce high-temperature steam. Its mathematical model is expressed as: （2）； In the formula, is the output thermal power of the gas boiler; is the natural gas consumption of gas boilers; is the heat generation efficiency of the gas boiler; The mathematical model of air conditioning is expressed as: （3）； In the formula, is the cooling power of the electric refrigerator; is the power consumption of the electric refrigerator; is the energy efficiency ratio of the electric refrigerator; The energy conversion equipment includes a steam-driven device, an absorption refrigerator, and an AC / DC converter. The steam-driven device is driven by medium-grade hot steam and low-grade hot steam is recovered. Its mathematical model is expressed as follows: （4）； In the formula, Low-grade thermal power recovered for steam-driven equipment; is the steam load; Heat recovery efficiency for steam-driven equipment; The mathematical model of the absorption refrigerator is expressed as: （5）； In the formula, is the cooling power of the absorption chiller; is the heat consumption power of the absorption refrigerator; is the energy efficiency ratio of the absorption chiller; The mathematical model of the AC-DC converter is expressed as: （6）； In the formula, and are the AC side input power and DC side output power of the AC / DC converter respectively; and are the conversion efficiencies from AC to DC and from DC to AC, respectively; The energy storage device includes an electric storage device and an ice storage device, wherein the mathematical model of the electric storage device is expressed as: （7）； In the formula, and are the battery power at time t and time t+1 respectively; and are the charging power and discharging power of the battery at time t respectively; is the self-consumption coefficient of the battery; and are the charging efficiency and discharging efficiency of the battery respectively; The mathematical model of the ice storage device is expressed as: （8）； In the formula, , and are the cooling powers of the ice storage device, refrigerator and ice storage tank respectively; , and are the power consumption of the ice storage device, the refrigerator and the ice storage tank respectively; is the energy efficiency ratio of the refrigerator; and are the ice storage capacity of the ice storage tank at time t and at time t+1 respectively; is the self-consumption coefficient of the ice storage tank; and They are the ice making energy efficiency ratio and refrigeration efficiency of the ice storage tank respectively; is the power consumption of the ice storage tank at time t; is the cooling power of the ice storage tank at time t.

4. The multi-microgrid multi-level demand response method considering fraudulent behavior according to claim 1, characterized in that: In S3, the objective function of the low-carbon economic optimization operation model of microgrid users is: （9）； In the formula, is the daily operating cost of user i; is the energy cost of user i; Equipment operation and maintenance costs for user i; The cost of starting and stopping the equipment of user i; is the interruption load cost of user i; is the carbon trading cost of user i; is the fraud cost of user i; T is time; in, The calculation formula is: （10）； In the formula, is the electricity price at time t; is the price of natural gas; is the price of hot steam; is the power purchased by user i at time t; is the gas purchasing power of user i at time t; is the purchased heat power of user i at time t; The calculation formula is: （11）； In the formula, The operation and maintenance cost per unit output power of the j-th device of user i; is the output power of the jth device of user i at time t; The calculation formula is: （12）； In the formula, The startup and shutdown costs of the jth device of user i; and are the status of the j-th device of user i at time t and time t-1 respectively. The status value is 1 for power on and 0 for shutdown; The calculation formula is: （13）； In the formula, is the interruption load of user i at time t; For peak load time; and is the cost coefficient corresponding to the interruptible load; The calculation formula is: （14）； In the formula, is the actual carbon emissions of user i; is the initial carbon emission quota of user i; is the unit carbon emission trading price; The calculation formula is: （15）； In the formula, For fraud factors; is the cost difference before and after user i participates in the response interaction.

5. A multi-microgrid multi-level demand response method considering fraudulent behavior according to claim 4, characterized in that: In S3, the constraints of the low-carbon economic optimization operation model of microgrid users are: Power balance constraints: The operation of the user's integrated energy system must meet the power balance constraints of electricity, heat, cooling and gas, specifically: （16）； （17）； （18）； （19）； （20）； In the formula, , , , are the AC load, DC load, heating load and cooling load of user i at time t respectively; Produce power for photovoltaics; is the output power of the jth gas turbine of user i at time t; is the power consumption of the jth ice storage tank of user i at time t; is the AC output power of the jth device of user i at time t; is the AC conversion power of the jth device of user i at time t; is the DC conversion power of the jth device of user i at time t; is the battery charging power of the j-th device of user i at time t; is the battery discharge power of the jth device of user i at time t; is the thermal power of the output of the j-th device of user i at time t; is the output low-quality thermal power of the j-th device of user i at time t; is the output thermal power of the jth device of user i at time t; is the low-grade thermal power recovered by the j-th device of user i at time t; is the steam load of the j-th device of user i at time t; is the heat consumption power of the jth device of user i at time t; is the cooling power of the electric refrigerator of the j-th device of user i at time t; is the cooling power of the ice storage device of the jth device of user i at time t; is the cooling power of the absorption chiller of the jth device of user i at time t; Equipment operation constraints, where the equipment itself has power upper and lower limit constraints and ramp rate constraints: （21）； （22）； In the formula, , are the upper and lower limits of the output power of the j-th device of user i, respectively; is the maximum climbing rate of the j-th device of user i; is the output power of the jth device of user i at time t-1; The energy storage capacity constraint of the energy storage device is: （23）； （24）； In the formula, and They are the highest and lowest power of the battery respectively; and They are the upper and lower limits of ice storage capacity of the ice storage tank respectively; is the battery capacity of user i at the jth time t; is the ice storage capacity of the j-th ice storage tank of user i at time t; and are the starting and ending power of the j-th battery of user i respectively; and are the starting ice storage capacity and ending ice storage capacity of the j-th ice storage tank of user i respectively; The constraints for energy storage devices to be charged and discharged at different times are: （25）； （26）； （27）； （28）； In the formula, is the maximum power of the battery; and is the introduced 0-1 variable; and are the maximum electrical power and maximum refrigeration power of the ice storage tank respectively; The fraud factor constraint is expressed as: （29）； In the formula, is the fraud factor limit; N is the number of users in the microgrid; It is the sum of all microgrid operating costs; It is the average cost of operating all microgrids.

6. A multi-microgrid multi-level demand response method considering fraudulent behavior according to claim 5, characterized in that: In the above S4, in the microgrid IDR market, the response quantity is traded as a commodity within the microgrid by utilizing the information interaction between the microgrid energy operator and each user, so as to achieve the goal of minimizing the overall cost of the microgrid on the basis of self-optimization by the user; after receiving the response quantity allocated by the microgrid energy operator, the user re-optimizes his own energy consumption plan considering the IDR, and reports the optimized IDR quotation and response quantity to the microgrid energy operator; The microgrid energy operator is responsible for the operation and clearing of the microgrid IDR market. Based on the quotation information and response capacity of each user, the response quantity allocation plan is determined with the goal of economic optimization, and the winning bid quantity and IDR market clearing price of each user are sent to each user. Specifically, the optimization clearing model of the microgrid energy operator is that when the power grid issues a peak-shaving demand, the microgrid energy operator allocates the peak-shaving amount that users should respond to in each time period, which can be expressed as: （30）； In the formula, is the response amount of the i-th user at time t; is the planned power purchase of the i-th user at time t without considering IDR; is the maximum available capacity of the i-th user at time t, which is calculated as follows: （31）； In the formula, is the peak load demand of the power grid at time t; is the maximum demand of the ith user; In the microgrid IDR market, the microgrid energy operator reasonably allocates the response volume at each peak-shaving moment according to the peak-shaving ratio and quotation reported by each user, determines the user's winning capacity and the IDR market clearing price. The decision-making goal of the microgrid energy operator is to minimize the total IDR cost of all users in the microgrid while ensuring that the total response volume of the microgrid meets the peak-shaving demand of the power grid. The objective function is: （32）； In the formula, is the total IDR cost of the microgrid supplier; is the set of supplier numbers, Quote the IDR for the i-th user; is the peak load ratio of the i-th user at time t; The peak load regulation amount of all users who win the bid should be equal to the peak load regulation demand of the power grid, that is, it meets the following requirements: （33）； The microgrid energy operator issues the clearing price at the maximum of all bidders’ bids. Users do not need to bear the peak load regulation of the power grid, so the power grid should compensate for this part of the response. After the microgrid IDR market settlement is completed, the microgrid energy operator is responsible for distributing the income obtained from the power grid to each user, which is expressed as: （34）； In the formula, The grid compensation fee allocated to the i-th user; The grid compensation fee obtained for the entire microgrid is calculated as follows: （35）； In the formula, Charges for peak load regulation services for the power grid; is the planned power purchase of the i-th user at time t after considering IDR; The user low-carbon economic optimization model considering IDR is that the user responds to the peak load demand of the power grid, and its operation optimization should consider the benefits in the IDR market. The objective function is: （36）； In the formula, The total cost after considering IDR for the i-th user; is the income obtained by the i-th user in the microgrid IDR market, and the calculation method is as follows: （37）； In the formula, It is the clearing price for the microgrid IDR market; In order to prevent users from exhausting the adjustable resources at a certain moment and to ensure that they have responsiveness at all peak-shaving moments, add constraints: （38）； In the formula, is a constant greater than 0, which means that the peak load ratio of user i at all peak load times is the same; In the microgrid IDR market, there are two types of users: suppliers and demanders. The actual IDR response of suppliers is greater than the maximum available capacity, and the excess can be sold to gain revenue. The actual IDR response of demanders is less than the maximum available capacity, and they need to purchase the insufficient IDR capacity. If it is greater than 1, it means that user i is a supplier in the microgrid IDR market. is positive, otherwise it is the demand side. is negative; After each user reports the response volume to the microgrid energy operator, the supplier also needs to declare its expected IDR price. , the IDR quotation of each supplier is declared according to the average cost of the IDR capacity actually sold, and the calculation method is as follows: （39）； In the formula, The peak load ratio When it is 1, the total cost of the i-th user after considering IDR; is the total cost of the i-th user after considering IDR.

7. A multi-microgrid multi-level demand response method considering fraudulent behavior according to claim 6, characterized in that: In S5, the non-cooperative game model in the microgrid IDR market is that in the microgrid IDR market, the final winning bid of each user is determined by the bid amount and the bid price submitted by all the users participating in the bidding. There is a game on the bid price among the users. The game result determines the probability of winning the bid in the IDR market in the park and then determines its interests. That is, the interests of users are affected by the decision-making behavior of other participants. Therefore, there is a non-cooperative game relationship between users in the microgrid IDR market. The game model can be characterized as follows: ; N b For game participants, that is, all users in the microgrid IDR market, ; S is the strategy available to all participants in the microgrid IDR market in the game. ; is the daily electricity purchase plan of user i, U is the utility function of all participants, ,in, is the utility function of user i, which is calculated as follows: （40）； In the non-cooperative game model in the microgrid IDR market, the existence of non-cooperative game equilibrium is explained as follows: The utility function of each participant in the microgrid IDR market is not only related to the strategy he chooses, but also depends on the strategies of other participants. Assuming that under a certain strategy combination, no participant can act alone to increase revenue, this strategy combination is called Nash equilibrium; In a multi-party non-cooperative game problem, if the strategy set of each participant is a non-empty compact convex subset of the Euclidean space, and the utility function is continuous quasi-concave in its strategy space, then the game problem has a Nash equilibrium; The game strategy in the non-cooperative game of the microgrid IDR market is the daily electricity purchase plan of each user, which needs to meet the constraints of the microgrid user low-carbon economic optimization operation model and the user low-carbon economic optimization model considering IDR. When user i participates in the microgrid IDR market bidding, there must be a corresponding daily electricity purchase plan, so the non-cooperative game strategy set is a non-empty compact convex subset; The utility function of user i is shown in formula (40): , If both are convex functions with respect to energy use strategies, then the linear superposition utility function of the two parts is a continuous quasi-concave function under the corresponding strategies; The solution process of the non-cooperative game model is: S51. The microgrid energy operator initially sets the peak-shaving service price of the power grid as the settlement price of the microgrid IDR market; the user reports the optimal peak-shaving ratio and IDR quotation under the current clearing price based on the user low-carbon economic optimization model considering IDR; S52, the microgrid energy operator summarizes the total peak load reported by all users and determines whether it is not greater than the peak load demand issued by the power grid, that is, whether formula (41) is established. If so, the IDR market is settled according to the peak load service price of the power grid, and the peak load ratio of each user is implemented according to its reported value, and the solution is completed; if not, continue to step 3, where formula (41) is: （41）； S53. The microgrid energy operator determines the IDR market clearing price and the total cost of fraud within the microgrid based on the microgrid energy operator's optimized clearing model; S54. The user reports the optimal peak load ratio and IDR quotation under the current clearing price based on the user low-carbon economic optimization model considering IDR; S55. The microgrid energy operator determines whether the formula is valid. If not, it returns to S53. If it is valid, the IDR market is settled according to the clearing price of the previous iteration, and the user's winning capacity is such that the total IDR amount = peak-shaving demand.

8. A multi-microgrid multi-level demand response method considering fraudulent behavior according to claim 7, characterized in that: In S6, the load aggregator alliance peak-shaving demand allocation strategy optimization model is: the load aggregator is referred to as an aggregator, and each aggregator forms an alliance. The goal is to minimize the total cost of all microgrids managed by the alliance and determine the optimal peak-shaving demand allocation strategy. The objective function is: （42）； In the formula, The total cost of the microgrid managed by the aggregator alliance after participating in IDR; S uni A collection of alliances formed by aggregators; is the total cost of the microgrid managed by aggregator m after participating in IDR; a microgrid managed by the alliance is a microgrid IDR market based on non-cooperative game; The optimization variable of the load aggregator alliance peak-shaving demand allocation strategy optimization model is the peak-shaving demand allocated to each microgrid. The total cost of each microgrid after participating in IDR is expressed as: （43）； The profit distribution mechanism is based on the distribution of power grid revenue, and the calculation formula is: （44）； In the formula, The revenue obtained from the grid by the microgrid managed by aggregator m after participating in IDR; is the cumulative value of the maximum demand of all users in the microgrid managed by aggregator m; is the cumulative value of the purchased power of all users of the microgrid managed by aggregator m at time t before alliance optimization; is the cumulative value of the purchased power of all users of the microgrid managed by aggregator m at time t after alliance optimization; is the total amount of peak demand transferred within the alliance; is the increased revenue of multi-microgrids due to cooperative game, and the calculation formula is: （45）； In the formula, is the total revenue of the alliance; is the profit of aggregator m before participating in the cooperative game; The response amount of users in the microgrid cannot be negative. Based on this premise, the upper and lower limits of the peak-shaving demand allocation are set, that is, the allocation of peak-shaving demand should meet the following requirements: （46）； In the formula, is the peak load demand allocated to microgrid m at time t; The PSO algorithm is used to solve the load aggregator alliance peak demand allocation strategy optimization model.

9. A multi-microgrid multi-level demand response method considering fraudulent behavior according to claim 8, characterized in that: In the above-mentioned S7, the process of solving the transaction mode is as follows: S71. Initialize fraud cost for each user , iteration number k=1, fraud factor =0.001, convergence accuracy =0.0001; S72. Each user calculates the fraud cost based on Fraud costs Upload to microgrid energy operators; S73. The microgrid energy application provider summarizes the fraud costs and feeds back the sum of the fraud costs to each user; S74, each user based on Update the fraud factor, where is the relaxation coefficient, and its value range is ; is the kth iteration ; is the fraud factor baseline value; S75, judge convergence, if it satisfies , output the total fraud cost, otherwise k=k+1, return to S72.

10. A multi-microgrid multi-level demand response method considering fraudulent behavior according to claim 9, characterized in that: In S8, the solution process of the multi-level distributed demand response optimization model is: S81. A coalition of load aggregators based on cooperative game constitutes an alliance optimization layer. Based on the alliance optimization layer, an initial particle group is randomly generated to distribute the peak demand allocation strategy to each microgrid. S82. Each microgrid completes the clearing of the microgrid IDR market based on non-cooperative game and the aggregation of the total fraud cost within the microgrid under the current peak-shaving demand allocation strategy, and returns the total cost of the microgrid to the load aggregator alliance; S83, the alliance optimization layer updates the historical optimal particle and the global optimal population according to the information returned by the microgrid, updates the position and speed of the particles, and generates a new particle swarm; S84. The alliance optimization layer sends the new particle swarm to the microgrid layer for cyclic iterative calculation. When the number of iterations reaches the specified value, the cycle stops and the last generation of population is output.

Images