Stackelberg intelligent building shared transaction system based on randomness and flexible load

Abstract

The invention discloses a Stackelberg intelligent building shared transaction system based on randomness and flexible load, and the system promotes the energy sharing between buildings and the effective utilization of distributed energy through the construction of a P2P (Peer-to-Peer) transaction mode. In consideration of energy consumption randomness and flexible load neglected by a traditional intelligent building optimization scheduling model, a probability model is established to simulate the randomness of photovoltaic power generation and power consumption demand; in the double-level optimization scheduling model, the upper layer maximizes the economic benefit of the load aggregator, and the lower layer improves the energy-saving benefit of the user; by introducing the Stackelberg game theory, existence and uniqueness of a model equilibrium solution are proved, a double-layer framework is designed to realize a dynamic interaction mechanism, economic and energy-saving benefits are comprehensively optimized, and the energy consumption rate is maximized while the economic benefits are ensured.

Description

Technical Field

[0001] The invention belongs to the technical field of power systems, and in particular relates to a Stackelberg intelligent building sharing transaction system based on randomness and flexible loads. Background Art

[0002] With the growth of global energy demand and the enhancement of environmental protection awareness, intelligent building energy management systems have gradually become an important means to improve energy efficiency and reduce carbon emissions. In existing technologies, most intelligent building energy management systems mainly rely on a single energy supply method, such as grid power supply, and fail to fully utilize the potential of distributed energy (such as photovoltaic power generation). In addition, existing systems still have shortcomings in dealing with the volatility of renewable energy and the flexible response of loads.

[0003] Smart buildings contain a variety of energy resources, such as photovoltaic power generation, energy storage systems, and electric vehicles. Although existing studies have explored the application of these energy resources in smart buildings, most of them focus on the optimization of a single energy resource and fail to fully consider the randomness of energy and the flexible response of loads. At the same time, traditional smart building optimization scheduling models fail to effectively integrate these resources, resulting in low system economic benefits and insufficient renewable energy consumption.

[0004] In recent years, the reform of the power market transaction model has introduced the P2P transaction model and the transaction model involving load aggregators to improve economic benefits and energy efficiency. However, the use of only the P2P transaction model is likely to lead to chaotic transaction order and difficult to achieve efficient energy utilization.

[0005] Demand response, as a means of changing the way electricity users use electricity through market price signals or incentive mechanisms, can reduce peak loads, balance power supply gaps, and improve the operating efficiency and stability of the power system. Although demand response is increasingly used in power systems, existing technologies are still insufficient in dealing with the randomness and flexibility of demand response.

[0006] For example, some existing studies have proposed smart building energy management methods based on the energy Internet, which improve the energy efficiency of the system through centralized scheduling and optimization. However, these methods usually ignore the energy sharing and trading between buildings, as well as the randomness of renewable energy generation and electricity demand, making it difficult for the system to achieve the expected results in actual operation. Summary of the invention

[0007] The purpose of the present invention is to solve the problems of low economic benefits, insufficient renewable energy consumption and lack of dynamic interaction mechanism in the prior art, and propose a Stackelberg intelligent building sharing transaction system based on randomness and flexible load. The system constructs a two-layer optimization scheduling model to comprehensively optimize economic benefits and energy-saving benefits, thereby realizing energy sharing among buildings and effective utilization of distributed energy.

[0008] The present invention is specifically a Stackelberg intelligent building sharing and trading system based on randomness and flexible loads. The intelligent building sharing and trading system includes multiple intelligent buildings, each of which is equipped with a photovoltaic power generation system, an energy storage system and an electric vehicle charging and discharging device. The power demand and supply information of each intelligent building is received through a trading platform, and energy trading is carried out.

[0009] The intelligent building sharing transaction system adopts a two-layer optimization scheduling model, the upper layer is the economic benefit maximization model of the load aggregator, and the lower layer is the energy-saving benefit improvement model of the user. The existence and uniqueness of the model equilibrium solution are ensured through the Stackelberg game theory.

[0010] The intelligent building sharing and trading system simulates the randomness of photovoltaic power generation and electricity demand by establishing a probability model, optimizes the balance of power supply and demand, and improves the stability and economic benefits of the system.

[0011] The intelligent building sharing and trading system promotes energy sharing among buildings and effective utilization of distributed energy through a P2P trading model, realizes a dynamic interaction mechanism, and ensures the stability and reliability of the system.

[0012] Compared with the existing technology, the beneficial effects are: a probabilistic model considering randomness and flexible loads is established, a two-layer optimization scheduling and Stackelberg game theory are adopted, a P2P transaction model is constructed, the economic benefits of the smart building energy management system are significantly improved, the renewable energy absorption capacity is enhanced, and the stability and reliability of the system are ensured. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 A probability distribution diagram of daily mileage of an electric vehicle of the present invention;

[0014] Figure 2 A probability distribution diagram of the start charging time of the electric vehicle of the present invention;

[0015] Figure 3 A simplified schematic diagram of the electric vehicle charging process of the present invention;

[0016] Figure 4 The P2P transaction process of the present invention;

[0017] Figure 5 is the total power consumption of the present invention;

[0018] Figure 6 is the photovoltaic output of the present invention;

[0019] Figure 7 The economic benefits of the present invention in different scenarios;

[0020] Figure 8 The optimized electricity price of the present invention;

[0021] Fig. 9 The user benefits before and after the optimization of the present invention are poor;

[0022] Fig.10 This is the building transaction volume for scenario 2 of the present invention;

[0023] Fig.11 This is the building transaction volume for scenario 3 of the present invention. DETAILED DESCRIPTION

[0024] The specific implementation of the Stackelberg intelligent building sharing transaction system based on randomness and flexible load of the present invention is described in detail below with reference to the accompanying drawings.

[0025] The intelligent building sharing trading system of the present invention includes multiple intelligent buildings, each of which is equipped with a photovoltaic power generation system, an energy storage system, and an electric vehicle charging and discharging device. The trading platform is used to receive the power demand and supply information of each intelligent building, optimize the power trading volume and price, and realize energy sharing between buildings and the effective use of distributed energy. The power demand and supply information of each intelligent building is received through the trading platform, and a two-layer optimization scheduling model is adopted. By establishing a probability model to simulate the randomness of photovoltaic power generation and power demand, the comprehensive optimization of economic benefits and energy-saving benefits is achieved.

[0026] The intelligent building sharing trading system optimizes the balance of power supply and demand by establishing a probability model to simulate the randomness of renewable energy generation, energy storage systems and electric vehicle charging and discharging, as follows:

[0027] The relationship between the power generation of a photovoltaic generator set and the overall efficiency and performance ratio of the photovoltaic system is shown below:

[0028]

[0029] Where: Photovoltaic unit is power generation; η sys It is the overall efficiency of the photovoltaic system, which is determined by factors such as the conversion efficiency of photovoltaic cells, the efficiency of components, and the loss of the system; A sysis the total area of ​​the PV system; G is the light intensity; PR is the performance ratio, which means the ratio between the actual power generation of the PV system and the theoretical maximum power generation; T is the power generation time; Losses is the loss of the system, which takes into account various loss factors in the system, including temperature loss of PV modules, inverter loss, cable and connector loss, shadow loss, fault loss, etc.

[0030] When simulating energy storage devices, it is necessary to consider the limits of their internal capacity and their state of charge during the charging and discharging phases:

[0031]

[0032] Where: The state of charge at time t and t-1 is S ESS,t , S ESS,t-1 ;μ ESS,t is the state indicator variable, with a value range of 0 and 1, representing the discharge and charge states respectively; η ESS-c , η ESS-d is the charging and discharging efficiency; the energy storage charging and discharging amount at time t is P ESS-c,t , P ESS-d,t ; E ESS is the capacity of the energy storage device.

[0033] Energy storage device power P ESS,t As shown below, P ESS,t >0 means the energy storage device is charged as a load, P ESS,t <0 indicates that the energy storage device is supplying power as a power supply response.

[0034]

[0035] In view of the uncertainty of electric vehicle charging and discharging behavior, Monte Carlo sampling is used for simulation:

[0036] The charging load of an electric vehicle is related to its daily mileage. When the sample is large enough, it can be considered that the amount of electricity consumed by the electric vehicle on that day is numerically equal to the amount of electricity required for charging on that day.

[0037] (1) Daily mileage probability distribution

[0038] Considering that electric vehicles and ordinary fuel vehicles have similar functions and user habits, this paper assumes that the daily mileage of electric vehicles is the same as that of ordinary fuel vehicles. By analyzing the travel pattern data of ordinary fuel vehicle owners, it is found that the daily mileage of electric vehicle owners follows a log-normal distribution, and its probability density function is as follows:

[0039]

[0040] In the formula, f s(x) is the probability density function of daily mileage; x is the daily mileage; mean μ s =3.3, variance σ s 2 =0.9.

[0041] It can be concluded that the probability distribution of daily mileage of electric vehicles is as follows: Figure 1 :

[0042] (2) Probability distribution of the starting charging time

[0043] The charging time of electric vehicles is concentrated between 17:00 and 23:00. The starting charging time t0 of electric vehicles follows a normal distribution, that is, t0~N(μ t ,σ t 2 ), its probability density function is as follows:

[0044]

[0045] In the formula, according to the actual time when the electric vehicle is connected to the grid, the average value μ is taken s =17, variance σ2s=3.5.

[0046] The probability distribution of the starting charging time of electric vehicles can be obtained Figure 2 shown.

[0047] (3) Charging power probability distribution

[0048] The charging power of electric vehicles depends on the type of battery and the charging method. Taking lithium batteries as an example, when an electric vehicle is connected to a charging pile, the battery begins to charge. The first stage is the beginning stage, when the electric vehicle has just been connected to the power grid, and the voltage and current provided by the power grid are relatively low, generally only 5-15% of the normal voltage. As time goes by, the charging process enters the second stage, the voltage and current gradually increase, the current growth rate is faster, and it rises to a certain value and remains constant, which belongs to the constant current stage. This stage is the normal charging state of the electric vehicle. The third stage is the final stage of charging, the voltage tends to be flat as the current decreases, and the current decreases rapidly under the constant voltage state until the charging is completed.

[0049] The charging method of the centralized access mode mostly adopts the DC fast charging method, which can fully charge the battery of the electric vehicle in a short time. Its main features include high power and high efficiency. In terms of power, the power of the DC fast charging charger is usually 50kW or above, which is much higher than the power of the AC slow charging charger. In terms of efficiency, the DC fast charging charger has a high conversion efficiency and can convert AC power into DC power for charging, reducing energy loss. This article uses the charging efficiency of the DC charging pile under full load conditions without efficiency optimization.

[0050] In the charging process of electric vehicles, the second stage occupies almost all the time of the entire charging process. Therefore, in the modeling process, the impact of the first and third stages on the entire charging process is ignored. When calculating, it is simplified into a horizontal straight line. The charging power of electric vehicles is as follows: Figure 3 As shown:

[0051] In the figure, the dotted line is the actual charging line, the solid line is the simplified charging power, and the simplified charging efficiency is 93%.

[0052] (4) Probability distribution of battery capacity

[0053] At present, there are certain differences in the battery capacity of electric vehicles. However, according to statistical data, it can be concluded that the battery capacity is roughly distributed within 20-30kWh. For the convenience of calculation, a uniform distribution is used to describe different battery capacities. The probability density is as follows:

[0054]

[0055] In the formula, f c (x) is the probability density of battery capacity, and x is the battery capacity.

[0056] (5) Electric vehicle charging and discharging model.

[0057] The battery state of charge in the electric vehicle charging and discharging model is shown below.

[0058]

[0059] Where: The battery state of charge at time t is S EV,t ;μ EV,t is a state indicator variable, which takes the value of 0 or 1, representing the discharge and charging states of the electric vehicle respectively; η EV-c , η EV-d are the charge and discharge efficiency respectively; P EV-c,t , P EV-d,t E are the charge and discharge amounts of the electric vehicle at time t; EV is the battery capacity.

[0060] The load model of the smart building is as follows:

[0061]

[0062]

[0063]

[0064] In the formula, is the total load of building h at time t; The common equipment load of the building; is the translatable load; P c u is a flexible load; It is the energy storage battery load; is the electric vehicle load; k tran is the translation coefficient; Adjust the upper limit for the flexible load of building h.

[0065] The establishment of the objective function is based on the master-slave game model, and a two-level master-slave game model is established. The upper leader is the load aggregator, and the lower followers are the users, namely buyers and sellers.

[0066] The upper function is the economic benefit, that is, the average economic benefit of the load aggregator. It reflects the average economic benefit of the load aggregator in the transaction by calculating the sum of the benefits from transactions with users and transactions with the power grid, as shown below.

[0067]

[0068] Where: P b P is the price at which the load aggregator sells electricity to the electricity purchaser. s The price of electricity purchased by the load aggregator from the electricity seller; buy The amount of electricity purchased by the electricity purchasing user, q sell The amount of electricity sold to electricity users; buy is the power purchase price of the grid, sell For the on-grid electricity price.

[0069] The lower function uses a logarithmic function to describe the energy-saving benefits of electricity sellers and buyers, as shown below.

[0070]

[0071]

[0072] Where: M n is the overall electricity efficiency of user n; q n For electricity consumption; k is the amount of electricity that the user can freely use during the current period; n is the benefit parameter of user n, which is always positive; the energy-saving benefit of electricity purchasing users is negative.

[0073] Constraints:

[0074] Electric vehicle charging and discharging constraints:

[0075] When building a charging and discharging management system for electric vehicles, various parameters must be precisely controlled to ensure efficient and safe operation of electric vehicles. In particular, the state of charge of an electric vehicle when it leaves a charging station is one of the key parameters to be monitored. For the i-th electric vehicle, its state of charge when it leaves a charging station should meet the following conditions:

[0076] S OC,i,min ≤S OC,i ≤S OC,i,max

[0077] Where: S oc,i is the charge state of the i-th electric vehicle when it leaves the charging pile; S oc,i,min , S oc,i,max are the lower and upper limits of the final state of charge of the i-th electric vehicle, respectively. This limit ensures that the electric vehicle has enough power to perform the scheduled trip while avoiding potential damage caused by overcharging the battery.

[0078] At any point in the life cycle of an electric vehicle, the state of charge must change according to the vehicle manufacturer's specifications and battery protection strategy, as shown below.

[0079] ε p S OC ≤S OC,i,t ≤S OC,i,max

[0080] Where: S oc,i,t is the state of charge of the ith electric vehicle at time t, ε p is the minimum power factor.

[0081] The charging and discharging power of electric vehicles is subject to constraints such as battery capacity and charging piles, and the charging and discharging powers are different, as shown below.

[0082] P EV,c,min ≤P EV.c ≤P EV,c,max

[0083] P EV,d,min ≤P EV.d ≤P EV,d,max

[0084] Where: P EV,c Charging power for electric vehicles, P EV,c,max , P EV,c,min are the upper and lower limits of the charging power respectively. EV,d is the discharge power of the electric vehicle, P EV,d,max , P EV,d,min are the upper and lower limits of the discharge power respectively.

[0085] Energy storage equipment constraints

[0086]

[0087] Where: It is the minimum and maximum output power of the power storage device.

[0088] The state of charge of the energy storage device is subject to upper and lower limits, and the energy storage device cannot be charged and discharged at the same time. As shown below.

[0089]

[0090] Where: It is the upper and lower limits of the state of charge of the energy storage device.

[0091] Photovoltaic power generation constraints

[0092] The upper and lower constraints of photovoltaic power generation are shown below.

[0093]

[0094] In the formula, are the lower and upper limits of photovoltaic power generation.

[0095]

[0096] Where: P pv,i,da (t) is the day-ahead predicted output of the i-th basic photovoltaic system; P pv,i,ba (t) is the basic output of the i-th basic photovoltaic system; B pv1,i (t), B pv2,i (t) is a binary variable, and when the value is 1, the uncertain variable reaches the boundary value of its interval; Δu pv,min (t), Δu pv,max (t) is the minimum and maximum fluctuation range of photovoltaic power generation output; Ф pv It is an uncertain parameter, which indicates the total number of time periods during which the photovoltaic power generation output and load power reach the minimum or maximum value in their fluctuation range within the scheduling period.

[0097] User power limit

[0098] For the electricity sales market, the total amount of electricity sold must be greater than or equal to the total amount of electricity purchased. At the same time, the total amount of electricity sold and the total amount of electricity purchased are at least zero. The difference between the total amount of electricity sold and the total amount of electricity purchased is the amount of electricity sold by the aggregator to the power grid, as shown below.

[0099]

[0100]

[0101] Similarly, for the electricity purchasing market, the total amount of electricity sold must be less than or equal to the total amount of electricity purchased. At the same time, the total amount of electricity sold and the total amount of electricity purchased are at least zero. The difference between the total amount of electricity sold and the total amount of electricity purchased is the amount of electricity purchased by the aggregator from the power grid, as shown in the formula below.

[0102]

[0103]

[0104] The user's transaction volume should be within the range of their own electricity consumption.

[0105]

[0106] Where: The minimum transaction amount for users; The maximum transaction volume of the user.

[0107] Electricity price constraints

[0108] The internal transaction electricity price is limited to between the on-grid electricity price and the grid electricity purchase price.

[0109]

[0110] Model solution:

[0111] P2P energy trading model

[0112] P2P transactions aim to maximize the buyer's revenue and reduce dependence on the power grid. To this end, a P2P energy trading mechanism based on double auctions is proposed, which can be divided into the following steps:

[0113] (1) Registration and declaration.

[0114] This article divides a day into 24 independent trading time periods, with one electricity transaction in each period. Before each hour of trading, participants register as buyers or sellers of electricity according to their own electricity needs, and provide the load aggregator with the expected amount of electricity purchased and sold. When distributed energy fails to meet energy demand, participants need to submit the amount of electricity purchased as buyers; when the output of distributed energy exceeds demand, they declare the amount of electricity sold as sellers.

[0115] (2) Optimization and Trading

[0116] a) The prosumer submits the desired electricity price and energy demand to the load aggregator, and the load aggregator provides the price P accordingly t .

[0117] b) The buyer observes the price P t And calculate the user benefit function to modify its own transaction needs and submit them to the load aggregator.

[0118] c) Repeat the above steps until the stopping condition is reached.

[0119] (3) Execution and Settlement

[0120] Fulfill the contract and verify whether the actual electricity consumption recorded by each participant’s smart meter is consistent with the electricity consumption specified in the contract.

[0121] Figure 4 For the P2P transaction process.

[0122] Stackelberg game model solution:

[0123] The load aggregator and the user can be defined as the leader and follower of the Stackelberg game respectively; the load aggregator provides the user with an electricity price, and the user provides the load aggregator with electricity consumption, which can be described as the process of the game. The game is shown below.

[0124] T={(N∪{LA}),Q n ,P b ,P s ,M n ,EWS}

[0125] Where N is the user set as followers, which decides the optimal power consumption according to the power price set by the load aggregator, LA is the load aggregator, which modifies the power price according to the power consumption adjusted by the user, and Q n The user's electricity consumption q n The strategy set ensures that the user's transaction volume does not exceed their own electricity consumption P b and P s are the electricity price strategy sets formulated by the load aggregators, and ensure that the transaction price range is within the power grid price range. n is the user's benefit function, and EWS is the load aggregator's benefit function.

[0126] As shown above, the goals of users and load aggregators are to maximize user benefits and load aggregator benefits respectively. When the maximum benefit is achieved, the load aggregator cannot obtain greater load aggregator benefits by modifying the electricity price, and the user cannot obtain greater user benefits by modifying his own electricity consumption. This point is the equilibrium point of the game. As shown below.

[0127]

[0128]

[0129] Where: Optimal electricity consumption for users; is the optimal electricity purchase price; p s *The optimal electricity selling price.

[0130] The Stackelberg game satisfies the following three requirements, which are sufficient conditions for the existence of a unique equilibrium solution of the game.

[0131] 1) The strategy set of the upper leader and the lower follower is a non-empty compact convex set.

[0132] 2) Once the leader’s strategy is determined, all followers have only the corresponding optimal solution.

[0133] 3) When all followers’ measurements are determined, the leader has only one optimal solution.

[0134] For electricity buyers, when the electricity consumption reaches the upper limit, further price reduction will not cause the user's electricity consumption to continue to increase. Based on this, the corresponding price can be obtained. Find all electricity buyers And sort them; similarly, the corresponding electricity price set can also be obtained in the case of electricity sales.

[0135]

[0136]

[0137] Where: Ω b is the lowest price set for electricity buyers; Ω s is the highest price set for electricity sellers. The above formula divides the feasible domain into N B ×N S It can be seen that the strategies of the game participants meet the requirements of non-empty compact convex sets.

[0138] Proof of the unique optimal solution of the lower objective function of the Stackelberg game:

[0139] Leadership Strategies and After determination, take the first derivative of the lower objective function and we can get the following formula.

[0140]

[0141]

[0142] From the above formula, we can see that at p s * and After confirmation, there is a unique q n * Correspondingly, we can find the second-order derivative of .

[0143]

[0144]

[0145] The second-order derivative is always negative, so is the minimum point, the optimal strategy can be expressed as:

[0146]

[0147] Therefore, when the leader's strategy and Once determined, the follower has one and only one Corresponding to it.

[0148] Proof of the unique optimal solution of the upper objective function of the Stackelberg game:

[0149] All electricity purchasing users constitute a set B1 and N B1 =|B1|; all electricity users constitute the set S1 and N S1 =|S1|. According to the electricity purchasing user set B and the electricity selling user set S, we can know that the amount of electricity purchased by all electricity purchasing users is q buy And all electricity sales collection q sell As shown below.

[0150]

[0151]

[0152] The function of the upper objective function with respect to price is obtained, and the second-order derivative thereof is obtained:

[0153]

[0154]

[0155] From the above formula, it can be seen that EWS is a strictly convex function of electricity price, so the optimal solution is unique in both intervals.

[0156] Case analysis:

[0157] A case analysis was conducted using three buildings with a total of 30 households with photovoltaic power generation capacity. The simulation was performed from 7:00 to 19:00. The total power consumption and photovoltaic output of each user were as follows: Figure 5 , 6 As shown in the figure, due to the large amount of data, the original data of 5 typical users are shown.

[0158] This paper will compare the differences of two-level objective functions in three scenarios. The scenario settings are shown in Table 1.

[0159] Table 1 Scenario classification

[0160]

[0161] Scenario 1 simulates a basic energy trading model, in which load aggregators only focus on maximizing their own economic benefits, without considering optimizing prices or incentivizing users to save energy through game strategies. In this model, users' energy-saving incentives and comprehensive energy utilization efficiency are not considered. Scenario 1 simulates the operating mode of most current energy markets, emphasizing the maximization of economic benefits, but resulting in suboptimal resource allocation and energy waste. Scenario 2 introduces the Stackelberg game strategy, where load aggregators set electricity prices as leaders to optimize economic benefits, and users respond as followers. However, this scenario still does not take energy-saving benefits as the optimization goal. Strategic games are introduced in energy trading to adjust prices and supply and demand, but users' energy-saving behaviors are still not fully motivated or considered, resulting in suboptimal energy utilization efficiency. Scenario 3 not only uses Stackelberg games to optimize trading prices, but also considers promoting users' energy-saving benefits, balancing economic benefits and energy-saving benefits, by introducing Stackelberg game theory and considering the impact of energy-saving benefits on optimizing economic benefits and improving energy efficiency. Through the simulation of different scenarios, the effectiveness of Stackelberg game theory in optimizing price mechanisms and energy regulation is verified, and the importance of energy-saving benefits in promoting efficient energy utilization is considered in energy trading strategies.

[0162] Economic benefit analysis: The economic benefit of the load aggregator is taken as the upper objective function optimized in this paper. The upper objective function values ​​under different strategies are as follows: Figure 7 shown.

[0163] From the analysis of the figure, we can see that:

[0164] (1) In scenario 1, the grid electricity price is used as the transaction price, and the Stackelberg game is not adopted. It can be seen from the figure that the economic benefits are less than those in scenarios 2 and 3. This is because when the grid electricity price is used as the transaction price, the user's trading enthusiasm decreases, the amount of electricity that the aggregator can trade decreases, and the price of electricity purchased from users increases, so the economic benefits decrease accordingly;

[0165] (2) In scenario 2, economic benefits are considered and Stackelberg game is adopted, but energy-saving benefits are not considered. As can be seen from the figure, the economic benefits are the highest among the three scenarios. When energy-saving benefits are not considered, aggregators are more inclined to improve their own benefits and ignore the energy clearing situation.

[0166] (3) In scenario 3, economic benefits and energy-saving benefits are comprehensively considered. Compared with scenario 2, the economic benefits are slightly reduced, but energy transactions are more frequent, the energy clearance level is improved, and the overall benefits are improved.

[0167] In economic benefits, electricity price is one of the core contents of the strategy. The optimization of electricity price in scenario 3 is as follows: Figure 8 shown.

[0168] Depend on Figure 8 It can be seen that during the morning peak hours of 8 to 9 o'clock, as the user's electricity consumption increases, the load aggregator moderately increases the electricity selling price to encourage electricity users to provide more electricity. At noon from 10 to 12 o'clock, the electricity demand decreases, and the load aggregator reduces the electricity selling price to prevent oversupply. From 13 to 15 o'clock, the photovoltaic output is in excess, and the load aggregator maintains a relatively stable electricity selling price at this time to ensure the continuity of electricity supply, while promoting user consumption by lowering the purchase price of electricity. From 16 to 17 o'clock, the demand for electricity surges, and the load aggregator significantly increases the electricity selling price to promote electricity supply, and increases the purchase price of electricity to control electricity demand and ensure the stability of the entire power grid. This electricity price adjustment strategy in different time periods shows how load aggregators optimize the electricity selling and purchase prices according to the real-time changes in electricity supply and demand, and guide users to modify their own electricity consumption to achieve dynamic equilibrium in the electricity market.

[0169] Depend on Figure 9-11 It can be seen that: using the two-layer model based on Stackelberg game, the benefits of users are improved in most time periods. Changes in electricity prices affect the energy-saving benefits of users. For electricity buyers, the electricity purchase price decreases, and their energy-saving benefits increase from 13 to 15 o'clock. For electricity sellers, the electricity sales price increases, and the energy-saving benefits increase from 8, 10-13 and 15-17 o'clock. Fig.10 and Fig.11 The energy interactions in scenarios 2 and 3 are shown respectively. In scenario 2, only economic benefits are taken as the goal. In order to pursue higher profits, aggregators purchase electricity from users at a lower price and sell it at a higher price. Therefore, users are less willing to trade energy and the energy clearing level is insufficient. In scenario 3, the shape of the transaction power curve is similar to that in scenario 2, but the transaction volume is larger, and some economic benefits are abandoned. For example, at 10 and 11 o'clock, the electricity price is increased, and more electricity users are willing to sell electricity, which improves the energy consumption rate.

[0170] Through the example analysis in this section, the effectiveness of the two-layer intelligent building sharing business model based on Stackelberg game in practical applications is verified. By comparing the economic and energy-saving benefits under different strategies, the results show that when the economic and energy-saving benefits are comprehensively considered, the model can not only improve the economic benefits of load aggregators, but also increase the energy-saving benefits of users, thereby improving the overall energy consumption rate. In addition, the model proposed in this paper effectively encourages users to participate in energy trading by optimizing the electricity price strategy, increasing the liquidity of energy and the activity of the market.

[0171] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not intended to limit the present invention. A person skilled in the art should understand that the specific implementation of the present invention can be modified or replaced by equivalents, but these modifications or changes are within the scope of protection of the pending claims.

Claims

1. A Stackelberg intelligent building sharing transaction system based on randomness and flexible load, characterized in that: The smart building sharing and trading system includes multiple smart buildings, each of which is equipped with renewable energy power generation equipment, energy storage systems and electric vehicle charging and discharging devices. The power demand and supply information of each smart building is received through the trading platform, and a two-layer optimization scheduling model is adopted to simulate the randomness of photovoltaic power generation and electricity demand by establishing a probability model to achieve comprehensive optimization of economic benefits and energy-saving benefits.

2. The Stackelberg intelligent building sharing transaction system based on randomness and flexible load according to claim 1 is characterized in that: The intelligent building sharing transaction system promotes energy sharing among buildings and effective utilization of distributed energy through a transaction platform based on a P2P transaction model.

3. The Stackelberg intelligent building sharing transaction system based on randomness and flexible load according to claim 2 is characterized in that: The P2P transaction model includes the following steps: (1) Registration and declaration; A day is divided into 24 independent trading time periods, with one electricity transaction in each period. Before each hour of trading, participants register as electricity buyers or sellers based on their own electricity needs, and provide the load aggregator with the expected amount of electricity purchased and sold. When distributed energy fails to meet energy demand, participants need to submit the amount of electricity purchased as electricity buyers. When the output of distributed energy exceeds demand, participants declare the amount of electricity sold as electricity sellers. (2) Optimization and trading, including: (21) The prosumer submits his / her desired electricity price and energy demand to the load aggregator, who then provides a price P. t ; (22) Buyer observes price P t And calculate the user benefit function to modify its own transaction requirements and submit them to the load aggregator; (23) Repeat steps (21)-(22) until the stop condition is reached; (3) Execution and settlement: fulfilling the contract and verifying whether the actual electricity consumption recorded by each participant’s smart meter is consistent with the electricity consumption specified in the contract.

4. The Stackelberg intelligent building sharing transaction system based on randomness and flexible load according to claim 1 is characterized in that: The intelligent building sharing transaction system optimizes the balance of electricity supply and demand by establishing a probability model to simulate the randomness of renewable energy generation, energy storage systems and electric vehicle charging and discharging.

5. The Stackelberg intelligent building sharing transaction system based on randomness and flexible load according to claim 1 is characterized in that: The intelligent building sharing transaction system adopts a two-layer optimization scheduling model, including an economic benefit maximization model for upper-layer load aggregators and an energy-saving benefit improvement model for lower-layer users. The Stackelberg game theory is used to ensure the existence and uniqueness of the model equilibrium solution and realize a dynamic interaction mechanism.

6. The Stackelberg intelligent building sharing transaction system based on randomness and flexible load according to claim 1 is characterized in that: Stackelberg game model solution: The load aggregator and the user can be defined as the leader and follower of the Stackelberg game respectively; the load aggregator provides the user with an electricity price, and the user provides the load aggregator with electricity consumption, which is described as the process of the game. The game is as follows: T={(N∪{LA}),Q n ,P b ,P s ,M n ,EWS} Where N is the user set as followers, which decides the optimal power consumption according to the power price set by the load aggregator, LA is the load aggregator, which modifies the power price according to the power consumption adjusted by the user, and Q n The user's electricity consumption q n The strategy set ensures that the user's transaction volume does not exceed their own electricity consumption P b and P s are the electricity price strategy sets formulated by the load aggregators, and ensure that the transaction price range is within the power grid price range. n is the user's benefit function, and EWS is the load aggregator's benefit function.

7. The Stackelberg intelligent building sharing transaction system based on randomness and flexible load according to claim 6 is characterized in that: In the intelligent building sharing transaction system, the load aggregator adjusts the electricity price according to user feedback.

8. The Stackelberg intelligent building sharing transaction system based on randomness and flexible load according to claim 7 is characterized in that: The intelligent building sharing and trading system effectively encourages users to participate in energy trading by optimizing electricity price strategies, increases energy circulation and market activity, verifies the effectiveness and rationality of the model through case analysis, and improves energy utilization and system stability.

Images