Double-layer energy transaction method for intelligent building group containing shared energy storage and power distribution network

Abstract

The invention discloses a double-layer energy transaction method for an intelligent building group with shared energy storage and a power distribution network, and the method comprises the steps: constructing an energy transaction framework for leasing the intelligent building group with shared energy storage and the power distribution network based on a strategy of a mixed game theory; an income distribution mechanism based on Nash negotiation is introduced, and energy sharing and scheduling in the intelligent building group are optimized; through a multi-objective optimization model, cost minimization and energy supply and demand stability are balanced, and the adaptive capacity of the system to renewable energy fluctuation is enhanced; an improved particle swarm algorithm is combined with a multiple multiplier alternating direction algorithm, the calculation efficiency and prediction accuracy of the model are optimized, and the limitation of a traditional energy transaction strategy is overcome; finally, in the mixed game solving process, a strategy based on the improved multi-objective particle swarm optimization (IMOPSO) is applied to the mixed game solving process as an effective example.

Description

Technical Field

[0001] The present invention belongs to the field of integrated energy technology, and in particular relates to a double-layer energy trading method between an intelligent building group with shared energy storage and a distribution network. Background Art

[0002] Against the backdrop of rapid globalization and urbanization, energy consumption and environmental pollution issues facing urban buildings are becoming increasingly prominent. Smart buildings, as an important part of modern urban buildings, not only achieve efficient energy use, but also optimize energy management processes by integrating automation control, information and communication technology, and energy management systems. These buildings use advanced automation systems to monitor and adjust lighting, air conditioning, heating, and electricity consumption, aiming to improve energy efficiency and reduce environmental burden. Although smart buildings have achieved remarkable results in improving energy efficiency and promoting sustainable development, they usually operate independently in an isolated form and lack an efficient energy sharing and coordination mechanism. This isolated operation mode limits the possibility of energy complementarity between different buildings, especially in the face of unstable renewable energy supply, it is difficult to achieve optimal energy balancing and scheduling. In addition, the complexity of the dynamically changing power demand and supply within the smart building cluster increases the difficulty of energy management.

[0003] Energy sharing, as an innovative model to solve this problem, advocates that smart buildings complement each other's electricity through shared energy storage facilities to achieve maximum efficiency and cost-effectiveness in energy use. However, current energy sharing strategies often fail to fully consider the dynamics, complexity and continuity of energy transactions between buildings, and especially ignore the fairness and economy of transactions. This leads to multiple technical and management challenges for smart buildings when implementing energy sharing. Centralized energy sharing systems are usually difficult to meet the diverse energy needs within smart building clusters due to their inherent rigidity. The lack of flexibility and adaptability of such systems makes it impossible to effectively respond to rapid changes in energy demand. In addition, the issue of revenue distribution of energy sharing has not yet been properly resolved, and there is a lack of a fair and reasonable mechanism, which reduces the enthusiasm of smart building clusters to participate in energy sharing and hinders its widespread application.

[0004] There are a series of technical and management challenges in the practice of power sharing in smart building clusters. Key issues include how to accurately predict and respond to changes in energy demand within buildings, how to achieve real-time optimal scheduling of energy between buildings, and maximize economic benefits while ensuring energy supply security. Power sharing in smart building clusters also involves issues such as data security, privacy protection, and compatibility with existing power grid systems. To meet these challenges, it is urgent to develop and implement a new power sharing business model that supports efficient configuration and optimal scheduling of electricity within smart building clusters and ensures fairness and economic benefits of power trading. This will not only promote the sustainable use of energy, but will also drive further innovation in smart building technology and management.

[0005] There are still a series of technical and management challenges in the practice of power sharing in smart building clusters, such as how to accurately predict and respond to changes in energy demand within buildings, how to achieve real-time optimization and scheduling of energy between buildings, and how to maximize economic benefits while ensuring energy supply security. Summary of the invention

[0006] In order to solve the above problems, the present invention provides a two-layer energy trading method for a group of smart buildings with shared energy storage and a distribution network, which can cope with problems such as uncertainty in electricity supply and demand and market price fluctuations, and further promote the integrated development of smart buildings and smart grids.

[0007] The present invention is a two-layer energy trading method for a smart building group with shared energy storage and a distribution network. The energy trading framework is constructed based on hybrid game theory, and the Nash negotiation theory is combined to achieve revenue distribution. The multi-objective optimization model is used to balance system cost and supply and demand stability, and the improved particle swarm algorithm and multiple multiplier alternating direction algorithm are used to improve the model calculation efficiency and prediction accuracy. The specific implementation steps include:

[0008] Step (1) Conducting in-depth analysis on the operation data of the smart building cluster and the distribution network to determine the key influencing factors;

[0009] Step (2) formulating an energy dispatching strategy for multiple microgrids to lease shared energy storage, optimizing the use of energy storage and smoothing net load fluctuations;

[0010] Step (3) construct a master-slave game model between the distribution network and the intelligent building group, with the distribution network as the leader and the building group as the responder, to optimize the electricity price and the amount of electricity purchased and sold;

[0011] Step (4) Nash negotiation theory is used to distribute the benefits within the building group, and a cooperative game model is constructed to ensure that the benefits among the buildings are fair and reasonable;

[0012] Step (5) developing a multi-objective optimization model to balance cost and supply and demand;

[0013] Step (6) implements an efficient algorithm optimization model to ensure real-time response and adaptability.

[0014] Compared with the prior art, the beneficial effects are: by adopting an energy trading framework based on hybrid game theory and a profit distribution strategy of Nash negotiation, the energy flow between smart building groups and distribution networks can be effectively managed and optimized; in dealing with the volatility and uncertainty of renewable energy supply, the present invention demonstrates excellent performance; secondly, the introduced multi-objective optimization model greatly improves the adaptability and response speed of the system to different energy demand scenarios, enabling it to maintain efficient operation under changing market conditions; in addition, by using an improved particle swarm algorithm and a multiple multiplier alternating direction algorithm, the computational efficiency and prediction accuracy of the model are further improved, especially when dealing with complex energy distribution and load scheduling problems within smart building groups; the two-layer energy trading method of the smart building group with shared energy storage and the distribution network of the present invention not only enhances the energy self-sufficiency of smart buildings, but also significantly improves the economic benefits and environmental sustainability of the entire system, providing an innovative management solution for complex urban energy networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A framework diagram of a two-layer energy trading method between a smart building group with shared energy storage and a distribution network according to the present invention;

[0016] Figure 2 It is the basic framework structure of the mixed game of the present invention;

[0017] Figure 3 It is a flow chart of solving the mixed game of the present invention;

[0018] Figure 4 It is the frontier surface of Pareto of each building of the present invention. DETAILED DESCRIPTION

[0019] The specific implementation method of a double-layer energy trading method between a smart building group with shared energy storage and a distribution network of the present invention is described in detail below with reference to the accompanying drawings.

[0020] like Figure 1 As shown, the double-layer energy trading method between the intelligent building group and the distribution network of the present invention comprises the following steps:

[0021] Step (1) Conducting in-depth analysis on the operation data of the smart building cluster and the distribution network to determine the key influencing factors;

[0022] Step (2) Formulate an energy dispatch strategy for multiple microgrids to lease shared energy storage, optimize the use of energy storage and smooth out net load fluctuations.

[0023] Step (3) construct a master-slave game model between the distribution network and the intelligent building group, with the distribution network as the leader and the building group as the responder, to optimize the electricity price and the amount of electricity purchased and sold;

[0024] Step (4) Nash negotiation theory is used to distribute the benefits within the building group, and a cooperative game model is constructed to ensure that the benefits among the buildings are fair and reasonable;

[0025] Step (5) developing a multi-objective optimization model to balance cost and supply and demand;

[0026] Step (6) implements an efficient algorithm optimization model to ensure real-time response and adaptability.

[0027] First, a multi-objective optimization model for smart building clusters to lease shared energy storage to smooth out their own net load fluctuations is established to reflect the on-demand leasing of energy storage by smart building clusters. Secondly, a master-slave-cooperative hybrid game model of the distribution network with embedded cooperative game of smart building cluster alliance members and the master-slave game of the smart building cluster alliance is constructed to take into account the respective interests of the distribution network and the smart building cluster alliance members. That is, the proposed model is divided into two steps: the first step is that the smart building cluster leases shared energy storage to smooth out power fluctuations and optimize its own net load; the second step is the master-slave-cooperative hybrid game between the distribution network and the smart building cluster. Among them, the hybrid game framework of the second step is as follows: Figure 2 The calculation steps are as follows:

[0028] (1) Based on Nash negotiation theory, the model is decomposed into two sub-problems: minimizing the operating cost of the intelligent building group (P1) and allocating cooperative benefits (P2);

[0029] (2) Solve subproblem P1, with the distribution network as the main body and the smart building group as the slave to conduct a master-slave game, with the distribution network maximizing its own benefits and the smart building group minimizing its operating costs (P1). Substitute the optimal time-of-use electricity price of the distribution network, the purchase and sale of electricity by the smart building group, the unit output, and the electricity traded between members of the smart building group obtained when the game is balanced into subproblem P2;

[0030] (3) Solve subproblem P2 and bring the inter-building group transaction power and the optimal time-of-use electricity price of the distribution network obtained in the second step into the cooperative game model of the smart building group to achieve profit distribution; finally, calculate the transaction electricity price between each member of the smart building group.

[0031] The specific content of the energy scheduling strategy for the intelligent building group to lease shared energy storage in step (2) is:

[0032] The shared energy storage service fee is defined as follows: shared energy storage providers have two charging methods: charging by rental capacity and charging by traffic. Charging by rental capacity is based on the capacity of the energy storage equipment leased by the smart building group, and charging by traffic is based on the actual charging and discharging power used by the user. Since the cost of leasing shared energy storage is still relatively high, there is a contradiction between this and the effect of the smart building group leasing energy storage to smooth its net load power fluctuations. Therefore, this paper constructs a multi-objective optimization model to resolve this contradiction. Objective one is to minimize the mean square error of the net load of the smart building group, and objective two is to minimize the energy storage rental cost. Among them, objective one can reflect the effect of leasing energy storage in the smart building group to smooth power fluctuations by setting the minimum net load mean square error. Each smart building group can set weight values ​​for different objectives according to preference, and obtain the corresponding energy storage charging and discharging strategy, reflecting the idea of ​​leasing energy storage on demand and improving the utilization rate of energy storage.

[0033] Objective function 1: Net load mean square error I of the smart building group after energy storage charging and discharging i Minimum

[0034]

[0035] Where: P L,i is the load of the ith intelligent building group in time period t; P WT,i is the wind power output of the i-th smart building group in time period t; P PV,i is the photovoltaic output power of the i-th smart building group in time period t; P CH,i and P DIS,i are the shared energy storage charging and discharging power of the i-th smart building group; P ave,i It is the average value of the equivalent load of the i-th intelligent building group in time period t.

[0036] Objective function 2: Minimize energy storage rental costs

[0037] C i =M rent，i +M sever，i (2)

[0038] The cost of leasing shared energy storage consists of two parts:

[0039] M rent,i The cost calculation formula for leasing shared energy storage capacity for the i-th smart building group is as follows:

[0040]

[0041] Among them, α i and β iare the unit discharge capacity and charging capacity fees paid by the i-th smart building group for using the shared energy storage service in time period t, T is the total number of time steps in the optimization cycle, N is the number of buildings in the smart building; M sever,i is the cost of leasing charging and discharging power from the shared energy storage provider according to the traffic volume of the i-th smart building group, and the calculation formula is:

[0042]

[0043] Among them, γ i It is the charging and discharging cost per unit power of shared energy storage.

[0044] The specific method of constructing the master-slave game between the distribution network and the smart building group alliance in step (3) is:

[0045] Objective function:

[0046] Distribution network operators (DNOs) maximize their own benefits DNO With this goal in mind, electricity prices are dynamically set to guide smart building clusters (SBCs) to conduct electricity purchase and sales transactions.

[0047]

[0048] in: and are the electricity selling price and electricity purchasing price of DNO in time period t respectively; SBC，s，t and P SBC，b，t are the electricity sold and purchased from DNO by the smart building group in time period t; C grid The cost of purchasing electricity from the upstream grid for DNO is calculated as follows:

[0049]

[0050] in, and is the electricity purchase price of the upper power grid, P DNO，s，t and P DNO，b，t They are respectively the power sold or purchased by DNO to the superior power grid in time period t.

[0051] Constraints:

[0052] The electricity prices set by DNO must meet the following upper and lower limits:

[0053]

[0054] in, and It is the upper and lower limits of the electricity selling price. and In order to prevent DNO from raising the electricity sales price or lowering the electricity purchase price when maximizing its own interests, the electricity purchase and sale prices must meet the average value constraint:

[0055]

[0056] in, and are the average prices of electricity sales and electricity purchase, respectively.

[0057] The smart building group responds to the time-of-use electricity price set by the DNO and optimizes the alliance's power purchase and sales. The optimized power purchase and sales include demand response, unit output and P2P power transactions between smart building groups. First, each building smoothes its net load fluctuations through shared energy storage to generate an equivalent load curve:

[0058] P SBC，eq，t =P L，t -P WT，t -P PV，t +P CH，t -P DIS，t (9)

[0059] Among them, P L，t is the load of the intelligent building group in time period t, P WT，t and P PV，t are wind power and photovoltaic power generation, P CH，t and P DIS，t are the charging and discharging power of the energy storage respectively.

[0060] The operating cost optimization goals of the intelligent building group are as follows:

[0061]

[0062] Among them, C DNO，i is the electricity purchase and sales cost between building i and DNO; C sever，i is the energy storage rental cost of building i;

[0063] C P2P，i is the electricity transaction cost between building i and other buildings.

[0064] The specific method of constructing the cooperative game model in step (4) is:

[0065] In the energy management of smart building clusters, each smart building achieves energy optimization scheduling through shared energy storage, energy trading and other means. In order to fairly and reasonably distribute the benefits of each building in cooperation, it is necessary to use a cooperative game model, especially the Nash bargaining model. This model can ensure that the participating entities reach a win-win benefit distribution plan through negotiation, so that the benefits of each smart building are not lower than the level when operating independently, thereby prompting the smart building cluster to actively participate in cooperation and achieve the optimization of overall energy efficiency.

[0066] The Nash bargaining model not only considers the cooperation between building groups, but also introduces asymmetric factors, which can fairly distribute benefits according to the role of each smart building in energy contribution. Through this cooperative game mode, building groups can effectively improve energy efficiency, reduce operating costs, and promote the consumption and storage of renewable energy.

[0067] Sub-problem (P1): Minimize the cost of smart building clusters

[0068] In the first sub-problem of the cooperative game, the goal of the smart building cluster is to minimize the overall operating cost. The operating cost of the smart building cluster can be divided into three parts:

[0069] Cost of purchasing and selling electricity: Smart buildings balance electricity demand by purchasing or selling electricity from the distribution network. This part of the cost is directly affected by the time-of-use electricity price set by the distribution network.

[0070] Energy storage rental costs: Each smart building can rent shared energy storage resources to smooth load fluctuations. This cost depends on the rental capacity and the actual charging and discharging flow used.

[0071] Internal transaction costs within a building group: Each building in a smart building group can conduct peer-to-peer (P2P) energy transactions to reduce dependence on the external power grid by sharing electrical energy.

[0072] The objective function is as follows:

[0073]

[0074] Where: C SBC is the total operating cost of the intelligent building group; C DNO，i is the cost of building i purchasing and selling electricity from the distribution network. The specific formula is:

[0075]

[0076] in, and are the electricity sales and purchase prices of the distribution network in time period t, P SBC，b，t and P SBC，b，t are the electricity sold and purchased from the grid by the building in time period t. sever，i is the energy storage rental cost of building i, which is composed of the leased capacity and the charge and discharge flow, and the formula is:

[0077]

[0078] Among them, α i and β i are the unit discharge and charging costs when building i uses shared energy storage.

[0079] C P2P，i is the P2P electricity transaction cost between building i and other buildings, and the formula is:

[0080]

[0081] Among them, r i，j，k is the electricity transaction price between building i and building j in time period t, P i，j，t It is the transaction amount between the two.

[0082] In the second sub-problem of the cooperative game, the intelligent building group distributes revenue through an asymmetric Nash bargaining model. The asymmetric Nash bargaining model is a revenue distribution method that takes into account the contribution of each participant and can distribute revenue according to the contribution of each building in the energy sharing process, thereby achieving a fair and reasonable distribution mechanism.

[0083] In this model, each smart building determines its bargaining power in the game based on the amount of electricity it provides or obtains. Buildings that provide more electricity will receive higher returns, while buildings that obtain more electricity will receive relatively lower returns. The formula for calculating bargaining power is as follows:

[0084]

[0085] Among them, E offer，i The electrical energy provided to building i during the cooperation period; E accept，i The amount of electricity obtained by building i during the cooperation period;

[0086] E offer，max and E accept，max The maximum value of electrical energy supplied and taken from all buildings.

[0087] After determining the bargaining power of each building, the revenue distribution is maximized through the following Nash bargaining formula:

[0088]

[0089] Among them: U i The benefits of building i after the cooperation; The revenue of building i when it operates independently.

[0090] Finally, combined with the power transaction results in the P1 subproblem, the asymmetric Nash bargaining model calculates the revenue distribution among the smart building groups through a distributed solution algorithm. In this way, the building group can distribute revenue fairly and effectively through power sharing, P2P transactions, etc., to ensure the maximization of the interests of all participants.

[0091] In step (5), a multi-objective optimization model is developed to balance cost and supply and demand:

[0092] In the energy dispatching and trading between smart building clusters and distribution networks, the combination of master-slave game and cooperative game can help smart building clusters optimize their energy trading strategies, reduce costs and increase profits. This section describes in detail the solution process of the master-slave-cooperative hybrid game model after the smart building cluster leases shared energy storage.

[0093] The game strategy is divided into three steps: the first step is to solve the multi-objective optimization model of intelligent building group leasing shared energy storage; the second step is to solve the master-slave game model; and the third step is to solve the cooperative game model.

[0094] (51) Solution of multi-objective optimization model for intelligent building cluster leasing shared energy storage

[0095] In order to minimize the energy storage cost of smart building clusters and effectively smooth the load fluctuations, an improved multi-objective optimization method is proposed. The improved multi-objective particle swarm algorithm (IMOPSO) is used to solve the multi-objective optimization problem. The optimization objectives include goal 1: minimizing the rental energy storage cost of smart building clusters; goal 2: smoothing the net load fluctuations of smart building clusters.

[0096] In order to find a compromise solution in multi-objective optimization, this paper introduces fuzzy set decision theory and calculates the optimal compromise solution through the fuzzy membership function. The formula is as follows:

[0097]

[0098] Among them, J m is the objective function value, and are the minimum and maximum values ​​of the objective function respectively. Through this optimization method, the energy storage cost and load fluctuation in energy trading can be effectively balanced.

[0099] (52) Solution of the master-slave game model for subproblem P1

[0100] In the master-slave game model, the distribution network operator (DNO) is the leader of the game and sets time-of-use electricity prices to guide the smart building cluster (SBC) to purchase and sell electricity. The equivalent net load of the smart building cluster after energy storage is introduced into the master-slave game model to minimize the operating cost of the building cluster.

[0101] The goal of DNO is to maximize its own revenue through electricity price strategy. The optimized objective function is:

[0102]

[0103] The intelligent building group optimizes the purchase and sale of electricity according to the electricity price strategy, and determines the best purchase and sale strategy by solving the KKT conditions.

[0104] (53) Solution of the cooperative game model for subproblem P2

[0105] In the cooperative game stage, the intelligent building groups distribute the benefits through P2P power trading. In this stage, an asymmetric Nash bargaining model is used, which can fairly distribute the cooperative benefits according to the contribution of the building group in the power trading. In order to protect the privacy of each participant, the alternating direction multiplier method (ADMM) is used to solve the cooperative game model.

[0106] The augmented Lagrangian function of the cooperative game model is as follows:

[0107]

[0108] Among them, d i is the bargaining factor of the building group members, λ i is the Lagrange multiplier. Through the distributed solution algorithm, the intelligent building group can reach the equilibrium of the cooperative game at the optimal electricity trading price.

[0109] (54) Solution process of the master-slave-cooperation mixed game

[0110] like Figure 3 , the solution process of the entire mixed game model is as follows:

[0111] Initial parameter setting: Set the initial power purchase and sales, energy storage capacity and fuzzy decision weight for each smart building group.

[0112] Multi-objective optimization solution: Use the IMOPSO algorithm to optimize the energy storage cost and load fluctuation smoothing effect of smart building groups.

[0113] Master-slave game solution: Optimize the DNO's electricity price strategy based on the KKT condition, and achieve the equilibrium of the master-slave game through the optimal purchase and sales of electricity by the smart building group.

[0114] Cooperative game solution: The revenue distribution of P2P electricity trading is carried out through the asymmetric Nash bargaining model and ADMM algorithm, and finally the optimal cooperation strategy between smart building groups is achieved.

[0115] Example

[0116] The present invention is further described below by means of an embodiment.

[0117] MATLAB2019b is used for simulation verification. The scheduling cycle is 24h and the time interval is 1h. The parameters used in the model are solved by the IMOPSO algorithm to solve the multi-objective optimization model of microgrid leasing shared energy storage. Set ω1=0.4 and ω2=0.6, and select the optimal compromise solution for the multi-objective optimization model of leasing shared energy storage to obtain the energy storage charging and discharging power strategy of each MG under the optimal compromise solution. Among them, the multi-objective Pareto frontier of each MG is as follows: Figure 4 shown.

[0118] In addition, Table 1 compares the multi-objective optimization results of each building. It can be seen that after the shared energy storage participated in the scheduling optimization, the net load mean square error of Building 1, Building 2 and Building 3 decreased by 70.5%, 57.9% and 77.0%, respectively. This further shows that leasing shared energy storage can effectively smooth the net load fluctuations of the building and promote the on-site consumption of renewable energy.

[0119] Table 1

[0120]

[0121] In order to verify the effectiveness of the strategy proposed in this paper in reducing the operating costs of the building group and improving the income of alliance members, Table 2 compares the income and cost results of Scheme 1 and Scheme 2. It can be found from Table 2 that compared with Scheme 2, since Scheme 1 takes cooperative game into consideration, the total income of the building group increased by 2318.4 yuan, and the income of DNO decreased by 2452.3 yuan. This is because MMA members can achieve mutual assistance in electricity through P2P electricity trading, reduce their dependence on DNO, and the income of DNO also decreases. Therefore, the income of building groups can be effectively improved through cooperative game. In addition, the results in Table 2 also show that after adopting the strategy proposed in this paper, the income increased by 688.3, ​​995.5 and 634.6 yuan respectively; the income improvement ratio was 11.99%, 35.03% and 7.08% respectively. Among them, Building 2 has the largest contribution to providing electricity in the process of P2P electricity trading, so its negotiation value is higher and the income improvement is the best. The results also show that after participating in P2P electricity sharing, the interests of each building are improved. This is because the introduction of asymmetric Nash bargaining theory ensures that the total cost of the building group is reduced while improving the benefits of each building. In addition, each building can achieve fair distribution of cooperative benefits based on its contribution to providing electricity in P2P electricity trading.

[0122] Table 2 Operation results under different schemes

[0123]

[0124] 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 two-layer energy trading method between a smart building group with shared energy storage and a distribution network, characterized in that: The double-layer energy trading method between the intelligent building group and the distribution network comprises the following steps: Step (1) Conducting in-depth analysis on the operation data of the smart building cluster and the distribution network to determine the key influencing factors; Step (2) formulating an energy dispatching strategy for multiple microgrids to lease shared energy storage, optimizing the use of energy storage and smoothing net load fluctuations; Step (3) construct a master-slave game model between the distribution network and the intelligent building group, with the distribution network as the leader and the building group as the responder, to optimize the electricity price and the amount of electricity purchased and sold; Step (4) Nash negotiation theory is used to distribute the benefits within the building group, and a cooperative game model is constructed to ensure that the benefits among the buildings are fair and reasonable; Step (5) developing a multi-objective optimization model to balance cost and supply and demand; Step (6) implements an efficient algorithm optimization model to ensure real-time response and adaptability.

2. A two-layer energy trading method between a smart building group with shared energy storage and a distribution network according to claim 1, characterized in that: The specific content of the energy scheduling strategy for the intelligent building group to lease shared energy storage in step (2) is: The shared energy storage service fee is defined as follows: shared energy storage providers have two charging methods: charging by rental capacity and charging by traffic. Charging by rental capacity is based on the capacity of energy storage equipment leased by the smart building group, and charging by traffic is based on the actual charging and discharging power used by the user. Objective function 1: Net load mean square error I of the smart building group after energy storage charging and discharging i Minimum Where: P L,i is the load of the ith intelligent building group in time period t; P WT,i is the wind power output of the i-th smart building group in time period t; P PV,i is the photovoltaic output power of the i-th smart building group in time period t; P CH,i and P DIS,i are the shared energy storage charging and discharging power of the i-th smart building group; P ave,i is the average value of the equivalent load of the i-th intelligent building group in time period t; Objective function 2: Minimize energy storage rental costs C i =M rent，i +M sever，i (2) The cost of leasing shared energy storage consists of two parts: M rent,i The cost calculation formula for leasing shared energy storage capacity for the i-th smart building group is as follows: Among them, α i and β i are the unit discharge capacity and charging capacity fees paid by the i-th smart building group for using the shared energy storage service in time period t, T is the total number of time steps in the optimization cycle, N is the number of buildings in the smart building; M sever,i is the cost of leasing charging and discharging power from the shared energy storage provider according to the traffic volume of the i-th smart building group, and the calculation formula is: Among them, γ i It is the charging and discharging cost per unit power of shared energy storage.

3. A two-layer energy trading method between a smart building group with shared energy storage and a distribution network according to claim 2, characterized in that: The specific method of constructing the master-slave game between the distribution network and the smart building group alliance in step (3) is: Objective function: Distribution network operators (DNOs) maximize their own benefits DNO To achieve this goal, we dynamically set electricity prices to guide smart building clusters (SBCs) to conduct electricity purchase and sales transactions: in: and are the electricity selling price and electricity purchasing price of DNO in time period t respectively; SBC，s，t and P SBC，b，t are the electricity sold and purchased from DNO by the smart building group in time period t; C grid The cost of purchasing electricity from the upstream grid for DNO is calculated as follows: in, and is the electricity purchase price of the upper power grid, P DNO，s，t and P DNO，b，t are the power sold or purchased by DNO to the upper grid in time period t; Constraints: The electricity prices set by DNO must meet the following upper and lower limits: in, and It is the upper and lower limits of the electricity selling price. and are the upper and lower limits of the power purchase price; in order to prevent DNO from raising the power sales price or lowering the power purchase price when maximizing its own interests, the power purchase and sale prices must meet the average value constraint: in, and are the average prices of electricity sales and electricity purchase respectively; The smart building group responds to the time-of-use electricity price set by the DNO and optimizes the alliance's power purchase and sales; the optimized power purchase and sales include demand response, unit output and P2P power transactions between smart building groups; first, each building smoothes its net load fluctuations through shared energy storage to generate an equivalent load curve: P SBC，eq，t =P L，t -P WT，t -P PV，t +P CH，t -P DIS，t (9) Among them, P L，t is the load of the intelligent building group in time period t, P WT，t and P PV，t are wind power and photovoltaic power generation, P CH，t and P DIS，t are the charging and discharging power of energy storage respectively; The operating cost optimization goals of the intelligent building group are as follows: Among them, C DNO，i is the electricity purchase and sales cost between building i and DNO; C sever，i is the energy storage rental cost of building i; C P2P，i is the electricity transaction cost between building i and other buildings.

4. A two-layer energy trading method between a smart building group with shared energy storage and a distribution network according to claim 3, characterized in that: The specific method of constructing the cooperative game model in step (4) is: The cooperative game model adopts the Nash bargaining model; Sub-problem (P1): Minimize the cost of smart building clusters In the first sub-problem of the cooperative game, the goal of the smart building cluster is to minimize the overall operating cost; the operating cost of the smart building cluster is divided into three parts: the cost of purchasing and selling electricity, the cost of energy storage leasing, and the transaction cost within the building cluster; The objective function is as follows: Where: C SBC is the total operating cost of the intelligent building group; C DNO，i is the cost of building i purchasing and selling electricity from the distribution network. The specific formula is: in, and are the electricity sales and purchase prices of the distribution network in time period t, P SBC，b，t and P SBC，b，t are the amount of electricity sold and purchased from the grid by the building in time period t. sever，i is the energy storage rental cost of building i, which is composed of the leased capacity and the charge and discharge flow, and the formula is: Among them, α i and β i are the unit discharge and charging costs when building i uses shared energy storage, respectively; C P2P，i is the P2P electricity transaction cost between building i and other buildings, and the formula is: Among them, r i，j，t is the electricity transaction price between building i and building j in time period t, P i，j，t is the transaction amount between the two; In the second sub-problem of the cooperative game, the intelligent building group distributes revenue through an asymmetric Nash bargaining model. In this model, each intelligent building determines its bargaining power in the game based on the electricity it provides or obtains. The calculation formula of bargaining power is as follows: Among them, E offer，i The electrical energy provided to building i during the cooperation period; E accept，i The amount of electricity obtained by building i during the cooperation period; E offer，max and E accept，max Provide and obtain the maximum amount of electrical energy in all buildings; After determining the bargaining power of each building, the revenue distribution is maximized through the following Nash bargaining formula: Among them: U i The benefits of building i after the cooperation; The revenue of building i when it operates independently; Finally, combined with the power transaction results in the P1 subproblem, the asymmetric Nash bargaining model calculates the profit distribution among the smart building groups through a distributed solution algorithm; in this way, the profit distribution within the building group is carried out fairly and effectively through power sharing, P2P transactions and other means, ensuring the maximization of the interests of all participants.

5. A two-layer energy trading method between a smart building group with shared energy storage and a distribution network according to claim 4, characterized in that: In step (5), a multi-objective optimization model is developed to balance cost and supply and demand: (51) Solving the multi-objective optimization model for leasing shared energy storage in smart building clusters; (52) Solve the master-slave game model of sub-problem P1; (53) Solve the cooperative game model of subproblem P2; (54) The solution process of the master-slave-cooperation mixed game.

6. A two-layer energy trading method between a smart building group with shared energy storage and a distribution network according to claim 5, characterized in that: The specific method for solving the multi-objective optimization model of intelligent building group leasing shared energy storage in step (51) is: An improved multi-objective particle swarm algorithm is used to solve the multi-objective optimization problem. The optimization objectives include: objective 1: minimizing the rental energy storage cost of the smart building group; objective 2: smoothing the net load fluctuation of the smart building group; The fuzzy set decision theory is introduced, and the optimal compromise solution is calculated through the fuzzy membership function. The formula is as follows: Among them, J m is the objective function value, and are the minimum and maximum values ​​of the objective function, respectively.

7. A two-layer energy trading method between a smart building group with shared energy storage and a distribution network according to claim 6, characterized in that: The master-slave game model solution method for subproblem P1 in step (52) is: In the master-slave game model, the distribution network operator (DNO) is the leader of the game and sets time-of-use electricity prices to guide the smart building cluster (SBC) to purchase and sell electricity. The equivalent net load of the smart building cluster after being smoothed by energy storage is introduced into the master-slave game model to minimize the operating cost of the building cluster. The goal of DNO is to maximize its own revenue through electricity price strategy. The optimized objective function is: The intelligent building group optimizes the purchase and sale of electricity according to the electricity price strategy, and determines the best purchase and sale strategy by solving the KKT conditions.

8. A two-layer energy trading method between a smart building group with shared energy storage and a distribution network according to claim 7, characterized in that: The cooperative game model solution method for subproblem P2 in step (53) is: In the cooperative game stage, the intelligent building groups distribute the benefits through P2P power trading. In this stage, the asymmetric Nash bargaining model is adopted, which can fairly distribute the cooperative benefits according to the contribution of the building groups in the power trading. In order to protect the privacy of each participant, the alternating direction multiplier method is used to solve the cooperative game model. The augmented Lagrangian function of the cooperative game model is as follows: Among them, d i is the bargaining factor of the building group members, λ i is the Lagrange multiplier. Through the distributed solution algorithm, the intelligent building group can reach the equilibrium of the cooperative game at the optimal electricity trading price.

9. A two-layer energy trading method between a smart building group with shared energy storage and a distribution network according to claim 8, characterized in that: The solution process of the master-slave-cooperation mixed game in step (54) is as follows: Initial parameter setting: set the initial power purchase and sales, energy storage capacity and fuzzy decision weight for each intelligent building group; Multi-objective optimization solution: Use the IMOPSO algorithm to optimize the energy storage cost and load fluctuation smoothing effect of smart building clusters; Master-slave game solution: Optimize the DNO's electricity price strategy based on KKT conditions, and achieve the equilibrium of the master-slave game through the optimal purchase and sale of electricity by the smart building group; Cooperative game solution: The revenue distribution of P2P electricity trading is carried out through the asymmetric Nash bargaining model and ADMM algorithm, and finally the optimal cooperation strategy between smart building groups is achieved.

Images