P2P (Peer-to-Peer) electricity / heat energy interactive operation method and device for differentiated building producers

Abstract

The invention provides a P2P electricity / heat energy interaction operation method and device for differentiated building producers and diastereomers, and belongs to the technical field of energy, and the method comprises the following steps: S1, constructing a P2P electricity / heat energy interaction model for differentiated building producers and diastereomers; and S2, solving the P2P electricity / heat energy interaction model of the producer and the consumer of the differentiated building by adopting an asynchronous distributed interaction solving algorithm based on the ADMM. According to the P2P electricity / heat energy interactive operation method and device for the differentiated building producers, the energy consumption cost of the producers of all the buildings can be remarkably reduced through the interactive method, the solving efficiency can be improved, and the globally optimal solution is guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of energy technology, and in particular to a differentiated building prosumer P2P electricity / heat energy interactive operation method and device. Background Art

[0002] As the main energy-consuming object in the urban integrated energy system, building prosumers are in urgent need of reasonable and efficient energy management methods. P2P electric energy interaction can be used as a key means to achieve efficient utilization of energy resources of building prosumers. However, building prosumers under P2P electric energy interaction can only convert the electric energy obtained by interaction to meet their heat demand, which may limit the energy resource utilization efficiency and economic benefits of building prosumers. With the rapid development of urban integrated energy systems, coordinated optimization and complementary advantages between different energy systems can provide a physical system basis for the interaction of electric / thermal energy between building prosumers. Therefore, carrying out P2P electric / thermal energy interaction between building prosumers for urban integrated energy systems is of great significance for further exploring and utilizing the flexibility of building prosumers and improving the economy of individual building prosumers and the overall urban integrated energy system.

[0003] In the process of P2P electricity / heat energy interaction, building prosumers can make full use of the building's heat storage characteristics and the flexibility provided by its energy resources such as multi-energy conversion equipment to improve their economic benefits. At the same time, the differences between building prosumers will affect the results of P2P electricity / heat energy interaction, and it is urgent to scientifically consider them in P2P electricity / heat energy interaction. However, existing studies only consider the differences in energy resource configuration and load distribution among building prosumers, and lack in-depth discussion of the differentiated characteristics of building prosumers such as the building insulation performance caused by the building age, quality or aging degree, and the performance of communication and computing facilities. In fact, these two types of characteristics will directly affect the utilization efficiency of the building prosumer's own energy resources and the efficiency of solving the interaction between building prosumers. In addition, as independent interest decision-making subjects, each building prosumer has the need for information privacy protection. Existing studies generally use distributed interactive solution algorithms, but the differences in the performance of communication and computing facilities make it difficult for existing synchronous distributed interactive solution algorithms to ensure efficient solution. The reason is that after determining its interaction information, the building prosumer with better communication and computing facility performance needs to wait to receive interaction information from other building prosumers with poor facility performance, so that the solution time of each iteration is equal to the solution time of the building prosumer with the worst facility performance. To this end, some studies have proposed asynchronous distributed interactive solution algorithms to solve this problem, by allowing the building prosumer with better facility performance to use the interaction information of the previous iteration of the prosumer with poor facility performance to update its own interaction information. However, when the facility performance gap between building prosumers is too large, the building prosumer with better facility performance needs to wait for a long time for the interaction information of other prosumers, and use the previously obtained information multiple times to update its interaction information. This situation may cause the updated interaction information to shift to the suboptimal direction, and thus fail to ensure the optimal final interaction solution. Summary of the invention

[0004] The purpose of the present invention is to provide a differentiated building prosumer P2P electricity / heat energy interactive operation method and device, which can significantly reduce the energy consumption cost of each building prosumer, improve the solution efficiency, and ensure the global optimal solution.

[0005] To achieve the above object, the present invention provides a differentiated building prosumer P2P electricity / heat energy interactive operation method, comprising the following steps:

[0006] S1. Construct a P2P electricity / heat energy interaction model for differentiated building prosumers;

[0007] The objective function of the P2P electricity / heat energy interaction model of differentiated building prosumers is as follows:

[0008]

[0009] in, represents the charging and discharging cost of the energy storage battery of building prosumer i during period t; represents the heat storage and release cost of the thermal storage equipment of building prosumer i during period t; It represents the cost of buying and selling electricity from building prosumer i to the operator of the urban integrated energy system during period t; represents the cost of purchasing heat energy from building prosumer i to the operator of the urban integrated energy system during period t; i represents the building prosumer number; t represents the time interval; T represents the interaction period; I represents the set of building prosumers; represents the charging power of the energy storage battery of building prosumer i during period t; represents the discharge power of the energy storage battery of building prosumer i during period t; represents the unit charging and discharging cost of the energy storage battery of building prosumer i during period t; represents the thermal storage power of the thermal storage equipment of building prosumer i during period t; represents the heat release power of the thermal storage equipment of building prosumer i during period t; represents the unit heat storage and release cost of the thermal storage equipment of building prosumer i during period t; and They represent the power of electricity and heat purchased by building prosumer i from the operator of the urban integrated energy system during period t, respectively. A positive value indicates electricity purchase, and a negative value indicates electricity sale; and They represent the prices of electricity and heat purchased by building prosumer i from the operator of the urban integrated energy system during period t;

[0010] The constraints of the P2P electricity / heat energy interaction model for differentiated building prosumers include the following six types:

[0011] Constraint 1: P2P electricity / heat interaction constraint;

[0012] Constraint 2: Constraints on the purchase and sale of energy from urban integrated energy system operators;

[0013] Constraint 3: Building thermal balance constraint;

[0014] Constraint 4: Energy storage battery, heat storage equipment, radiator and heat pump operation constraints;

[0015] Constraint 5: Electricity / heat power balance constraint of building prosumers;

[0016] Constraint 6: Distribution network constraints and heating network constraints;

[0017] S2. An asynchronous distributed interactive solution algorithm based on ADMM is used to solve the P2P electricity / heat energy interaction model of differentiated building prosumers, including the following sub-steps:

[0018] S21, setting the index k of each building prosumer to 1, and setting the initial values ​​of the coupling variables, auxiliary variables, Lagrange multipliers and step length of each building prosumer, using the ADMM algorithm, and decomposing the coupling variables to obtain the differentiated building prosumer P2P electricity / heat energy interaction sub-problem;

[0019] S22, each building prosumer uses the latest auxiliary variable values ​​and Lagrange multipliers to solve its subproblem in an asynchronous manner;

[0020] S23, executing the prediction step, updating the auxiliary variable value and Lagrange multiplier of the kth iteration;

[0021] S24, performing anomaly detection steps to evaluate the prediction results of the kth iteration;

[0022] When the prediction result does not cause an exception, proceed to step S25;

[0023] When the prediction result causes an exception, only the auxiliary variable value is updated, and k=k+1 is set, and then the process returns to step S22;

[0024] S25, when all the original residuals and the dual residuals are less than the convergence threshold, as shown in formula (2) and formula (3), the iteration stops; otherwise, return to step S22;

[0025]

[0026]

[0027] Among them, r primal represents the original residual; Indicates the kth i The coupling variables of the iteration; Indicates the kth i Auxiliary variable for the iteration; r dual represents the dual residual; ρ represents the step size; Indicates the kth i -1 Auxiliary variables for iteration; δ1 and δ2 both represent convergence thresholds.

[0028] Preferably, the P2P electrical / thermal energy interaction constraints are as follows:

[0029]

[0030]

[0031]

[0032]

[0033] in, and They represent the P2P electric energy and thermal energy interaction power between building prosumer i and building prosumer j during period t, respectively. A positive value indicates that energy is purchased from building prosumer j, and a negative value indicates that energy is sold to building prosumer j. and They represent the P2P electric and thermal energy interaction power between building prosumer j and building prosumer i during period t, respectively. A positive value indicates that energy is purchased from building prosumer i, and a negative value indicates that energy is sold to building prosumer i. and They represent the maximum values ​​of interactive electric energy sold and purchased by building prosumer i to building prosumer j respectively; and They represent the maximum interactive heat energy sold and purchased by building prosumer i to building prosumer j, respectively.

[0034] Formula (4) and Formula (5) represent the upper and lower limit constraints of the P2P electricity / heat energy interaction power of building prosumers, and Formula (6) and Formula (7) represent the P2P electricity / heat energy interaction power balance constraints between building prosumers.

[0035] Preferably, the constraints on energy purchase and sale to the urban integrated energy system operator are as follows:

[0036]

[0037]

[0038] in, and They represent the maximum power of electricity and heat purchased by building prosumer i from the operator of the urban integrated energy system.

[0039] Preferably, the building heat balance constraints are as follows:

[0040]

[0041]

[0042] Among them, 1 represents the reference node of the heating area, recorded as node 1; 2 and 3 represent the nodes of the two areas adjacent to the heating area, recorded as nodes 2 and 3; 4 and 5 represent the nodes of the two outdoor areas, recorded as nodes 4 and 5; w represents the node number; there is a window installed on the wall of node 5; the heat capacities of the four walls around the heating area of ​​building prosumer i are The thermal resistances are The wall temperatures during period t are The wall temperatures during the t+1 period are represents the temperature of node 1 of building prosumer i during period t, They represent the temperature of node 2, node 3, node 4, and node 5 of building prosumer i in period t respectively; r 1,2 、r 1,3 、r 1,4 、r 1,5 They represent the state of sunlight exposure to the wall. If it is exposed to sunlight, the value is 1, otherwise it is 0; α 1,2 , α 1,3 , α 1,4 , α 1,5 Respectively represent the heat absorption rate of the four walls; Respectively represent the surface areas of the four walls; They represent the solar radiation intensity received by the four walls of building prosumer i respectively; Δt represents the time difference; represents the indoor air heat capacity of node 1 of building prosumer i; T i out represents the outdoor temperature of building prosumer i; represents the indoor temperature of building prosumer i during period t; represents the indoor temperature of building prosumer i in period t+1; represents the temperature of the wall between the building prosumer i and w nodes during period t; represents the thermal resistance of windows of building prosumer i; represents the heat increment of the temperature control load in node 1 of building prosumer i during period t; represents the heat gain generated by indoor residents and equipment in node 1 of building prosumer i during period t; τ win represents the transmittance of the window; represents the area of ​​windows of building prosumer i; It represents the solar radiation intensity of the windows of building prosumer i during period t;

[0043]

[0044]

[0045] in, represents the heat load of building prosumer i at node 1 during period t; and They represent the upper and lower limits of the heat load of node 1 in building prosumer i during period t respectively; represents the indoor temperature of node 1 in building prosumer i during period t; and They represent the upper and lower limits of the temperature comfort range during period t respectively.

[0046] Preferably, the energy storage battery operation constraints include charge and discharge power constraints and SOC limit constraints:

[0047]

[0048]

[0049]

[0050]

[0051]

[0052]

[0053] in, and They represent the charging and discharging power of the energy storage battery of building prosumer i during period t; P i ba,chg,max and P i ba,dischg,max They represent the maximum charging and discharging power of the energy storage battery of building prosumer i during period t respectively; represents the SOC, i.e., the state of charge, of the energy storage battery of building prosumer i during period t; and They represent the upper and lower limits of the SOC of the energy storage battery of building prosumer i in period t respectively; represents the capacity of the energy storage battery of building prosumer i; and They represent the charging and discharging efficiency of the energy storage battery of building prosumer i respectively; represents the initial SOC of the energy storage battery of building prosumer i, indicating that the SOC of the energy storage battery remains unchanged at the initial period t0 of each day, and there are d periods every day; Indicates the SOC of the energy storage battery in time period t0+d; Represents the SOC of the energy storage battery in time period t-1;

[0054] The relationship between the heat storage capacity, heat storage power, heat release power and heat dissipation of the heat storage equipment is as follows:

[0055]

[0056] in, represents the heat storage of the heat storage equipment of building prosumer i in period t+1; represents the heat storage of the heat storage equipment of building prosumer i during period t; represents the thermal storage power of the thermal storage equipment of building prosumer i during period t; represents the heat release power of the heat storage equipment of building prosumer i during period t; ηTES Indicates the heat dissipation rate of the thermal storage device;

[0057] The operating constraints of thermal storage equipment are as follows:

[0058]

[0059]

[0060]

[0061]

[0062] in, and They represent the maximum heat storage and heat release power of the heat storage equipment of building prosumer i; W i TES,min and W i TES,max They represent the minimum and maximum heat storage capacity of the thermal storage equipment of building prosumer i respectively;

[0063] The heat sink operation constraints are as follows:

[0064]

[0065] Among them, c p represents the specific heat capacity of water; represents the water flow rate of building prosumer i in the radiator during period t; T s and T r Respectively represent the radiator water supply and return temperatures; represents the heat power transferred by the radiator to the prosumer i of the building during period t;

[0066] The power consumption of heat pump is as follows:

[0067]

[0068] in, represents the power consumption of the heat pump of building prosumer i during period t; represents the heating power of the heat pump of building prosumer i during period t; COP HP It represents the energy efficiency ratio of the heat pump;

[0069] The heat pump operation constraints are as follows:

[0070]

[0071] Among them, P i HP,max represents the maximum power consumption of the heat pump of building prosumer i; Represents the power consumption of the heat pump of building prosumer i during period t.

[0072] Preferably, the building prosumer electricity / heat power balance constraint is as follows:

[0073]

[0074]

[0075]

[0076] in, represents the total heat load of building prosumer i during period t; represents the power of the rigid electrical load of building prosumer i during period t; Indicates the actual photovoltaic power generation output; represents the actual photovoltaic power generation output of building prosumer i during period t; j represents the building prosumer number; represents the power of heat energy produced by the heat pump of building prosumer i during period t; represents the heat load of heating zone n in building prosumer i during period t; n represents the number of heating zones; N represents the total number of heating zones;

[0077] Formula (28) represents the electric power balance constraint of building prosumer i in period t; Formula (29) represents the thermal power balance constraint of building prosumer i in period t; Formula (30) represents that the total heat load of building prosumer i is equal to the sum of the heat loads of all its heating areas;

[0078] The distribution network constraints use the linearized DistFlow model. The active power, reactive power and voltage of any branch are expressed as:

[0079]

[0080]

[0081]

[0082] 1-ε≤V m ≤1+ε (34);

[0083] Among them, P m+1 represents the active power of node m+1; P m represents the active power of node m; r f Indicates line resistance; Q m represents the reactive power of node m; represents the active power of the load at node m+1; Q m+1 represents the reactive power of node m+1; x f Indicates line reactance; represents the reactive power of the load at node m+1; Pm +iQ m represents the complex power injected by node m; V m Represents the voltage of node m; V m+1 represents the voltage of node m+1; represents the complex power of the load at node m+1; r f +ix f represents line impedance; V0 represents the branch root node voltage; ε represents the node voltage operating fluctuation range;

[0084] The heating network constraint adopts the heating network model, which is as follows:

[0085] Y HDN m pipe =m node (35);

[0086]

[0087]

[0088]

[0089]

[0090] Among them, Y HDN Represents the incidence matrix of the heating network; m pipe and m node Respectively represent each pipeline flow and node flow vector; and They represent the maximum and minimum water flow in pipeline l during period t respectively; N pipe Indicates the number of pipelines; p n,t represents the water pressure at node n during period t; ξ l represents the pipeline pressure coefficient; κ l , L l and d l They represent the friction coefficient, length and inner diameter of the pipe l respectively; ρ water Indicates water density; N node Indicates the number of nodes; and Respectively represent the maximum and minimum values ​​of the node pressure; represents the hot water flow in pipe l during period t; p n+1,t represents the water pressure at node n+1 during period t;

[0091] Formula (35) indicates that the flow rate injected into the node in the heating network is equal to the flow rate out of the node. At the same time, due to the short length and small scale of the heating network, the present invention ignores the heat loss in the heating network; Formula (36) indicates that the water flow rate in the pipeline needs to be constrained within a certain range to prevent pipeline vibration; Formula (37) indicates that the pipeline pressure drop caused by the friction of water flow along the pipeline is proportional to the square of the water flow rate; Formula (39) indicates that the node pressure needs to be controlled within a certain range to ensure the safety of the heating network;

[0092] Among them, formula (37) contains the square term of the variable Therefore, the piecewise linearization method is used to deal with it; by reducing the range of formula (36) Divide into Q segments, q∈Q and represent them on the abscissa, and the corresponding ordinate and slope are as follows:

[0093]

[0094] Then the nonlinear equation in formula (37) is processed piecewise and replaced by the sum of a set of linear equations, as shown in formula (41)-formula (43):

[0095]

[0096]

[0097]

[0098] in, is a 0-1 variable, indicating Whether the value is on the q segment.

[0099] Preferably, in step S22, each building prosumer uses the latest auxiliary variable value and Lagrange multiplier to solve its sub-problem in an asynchronous manner, and the specific operation is:

[0100] Determine the coupled variables in the building prosumer subproblem and then introduce auxiliary variables To ensure that the coupling variables are equal, thus forming consistency constraints between sub-problems:

[0101]

[0102]

[0103] The Lagrangian relaxation method is used to extend the above consistency constraint into the objective function (1):

[0104]

[0105] Among them, χ i Represents a local variable; xi represents the coupling variable, i.e., the P2P electricity / heat interaction power of building prosumer i; z i(j) Represents auxiliary variables; and They represent the P2P electric energy and thermal energy interaction power between building prosumer i and building prosumer j in the k+1th iteration period t; Cost i The same variables as in the objective function (1), Both are used to represent the P2P electric / thermal energy exchange coupling between differentiated building prosumers; ρ represents the step size; It represents the Lagrange multiplier corresponding to the consistency constraint in time period t, that is, the P2P electricity / heat energy interaction price.

[0106] Preferably, in step S23, a prediction step is performed, and the prediction step is performed to update the auxiliary variable value and the Lagrange multiplier of the kth iteration. The specific operations are as follows:

[0107] Assume that building prosumer i is in the kth i In the kth iteration, if the coupling variable value transmitted by building prosumer p is not received, building prosumer i will predict the value of the variable and use the kth iteration to predict the value of the variable. i -2nd and kth i -1 The coupling variable values ​​in the iteration are predicted using momentum extrapolation:

[0108]

[0109]

[0110] in, Indicates the kth i -2nd and kth i -1 The trend of the coupling variable value in the iteration; μ and η are weight factors; Indicates the kth i -3rd and kth i - Trend of coupling variable values ​​in 2 iterations; and Respectively represent the kth i -1 and k i - The coupling variables of building prosumer i and building prosumer p in the 2nd iteration; Indicates the kth i The predicted value of the coupled variable between building prosumer i and building prosumer p in the iteration;

[0111] It is worth noting that k in formula (47) and formula (48) i Need to be greater than 2; when k iWhen it is less than or equal to 2, the building prosumer will directly adopt the coupling variable value of the first iteration or the initial setting.

[0112] The predicted value of the coupling variable obtained based on formula (47) and formula (48) and the actual value of the coupled variable Calculate the kth i +1 Auxiliary variable values ​​and Lagrange multipliers for iteration:

[0113]

[0114]

[0115]

[0116]

[0117] In the formula, Indicates the kth i +1 iteration of the auxiliary variable of electric energy for building prosumer i in period t; Indicates the kth i The thermal energy auxiliary variable of building prosumer i in the iteration t period; and Respectively represent the kth i The P2P electric and thermal energy interaction power between building prosumer i and building prosumer a during the iteration t period; and Respectively represent the kth i The P2P electric and thermal energy interaction power between building prosumer i and building prosumer p during the iteration t period; and Respectively represent the kth i The Lagrange multiplier of the electrical and thermal energy of building prosumer i during the iteration t period; and Respectively represent the kth i -1 Lagrangian multipliers of electrical and thermal energy of building prosumer i and building prosumer a during period t of iteration t; and Respectively represent the kth i -1 Lagrange multipliers of electrical and thermal energy of building prosumer i and building prosumer p in time period t; and Respectively represent the kth i The auxiliary variables of electric energy and thermal energy of building prosumer i and building prosumer a in the iteration t period; and Respectively represent the kth i The auxiliary variables of electrical energy and thermal energy of building prosumer i and building prosumer p in the iteration t period;

[0118] Construct the building prosumer i in the kth i The sub-problem of +1 iteration is as follows:

[0119]

[0120] Where N a N represents the total number of building prosumer sellers; p Represents the total number of building prosumer buyers.

[0121] Preferably, in step S24, the abnormality detection step is as follows:

[0122] Based on the energy cost of the building prosumer i in the previous s iterations, the corresponding cost curve is obtained;

[0123] The wavelet threshold method and moving average method are used to reduce the noise of the cost curve until the noise-reduced curve and the original curve f i The difference between the original curve f i The ratio of is less than the threshold ε, as shown in formula (54) and formula (55):

[0124]

[0125]

[0126] in, All represent the Lagrange multipliers corresponding to the consistency constraints;

[0127] Based on the above difference, anomaly detection is performed on the predicted value in the k+1th iteration.

[0128] Specifically, the K-Means method is used to divide the difference between the denoised curve and the original curve results of the s iterations into ψ groups, and then the group that will cause anomalies is determined. i When the difference of an iteration belongs to a group that will cause an exception, the Lagrange multiplier of that iteration will not be updated, thus ensuring that the above asynchronous distributed solution algorithm generates the optimal interaction solution.

[0129] The present invention also provides a device for implementing the above-mentioned differentiated building prosumer P2P electricity / heat energy interactive operation method, including a differentiated building prosumer P2P electricity / heat energy interactive model construction module and a differentiated building prosumer P2P electricity / heat energy interactive model solving module; the differentiated building prosumer P2P electricity / heat energy interactive model solving module includes five submodules, namely, an initialization module, a sub-problem solving module, a prediction module, an anomaly detection module and a discrimination module;

[0130] Among them, the differentiated building prosumer P2P electricity / heat energy interaction model construction module is used to construct the differentiated building prosumer P2P electricity / heat energy interaction model;

[0131] The differentiated building prosumer P2P electricity / heat energy interaction model solving module is used to solve the differentiated building prosumer P2P electricity / heat energy interaction model through an asynchronous distributed interactive solving algorithm based on ADMM;

[0132] The initialization module is used to set the index of each building prosumer and the initial values ​​of the coupling variables, auxiliary variables, Lagrange multipliers and step length of each building prosumer. The ADMM algorithm is used to decompose the coupling variables to obtain the differentiated building prosumer P2P electricity / heat interaction sub-problems;

[0133] The sub-problem solving module is used to solve the sub-problems of each building prosumer in an asynchronous manner using the latest auxiliary variable values ​​and Lagrange multipliers;

[0134] The prediction module is used to perform the prediction step and update the auxiliary variable values ​​and Lagrange multipliers of the kth iteration;

[0135] The anomaly detection module is used to perform the anomaly detection step and evaluate the prediction results of the kth iteration;

[0136] The discriminant module is used to judge the original residual, dual residual and convergence threshold.

[0137] Therefore, the present invention adopts the above-mentioned differentiated building prosumer P2P electricity / heat energy interactive operation method and device. This interactive method can significantly reduce the energy consumption cost of each building prosumer, improve the solution efficiency, and ensure the global optimal solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0138] Figure 1 is the building RC network model; where Figure 1 (a) is a single wall RC network model; Figure 1 (b) is the RC network model of a single heating area;

[0139] Figure 2 It is a schematic diagram of the distribution network branch;

[0140] Figure 3 Solve the P2P electricity / heat energy interaction process between building prosumers based on synchronous and asynchronous distributed algorithms;

[0141] Figure 4 It is a flowchart of P2P electricity / heat energy interactive solution for building prosumers based on asynchronous distributed interactive solution algorithm;

[0142] Figure 5Schematic diagram of the structure of P2P electricity / heat energy interaction for differentiated building prosumers. DETAILED DESCRIPTION

[0143] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0144] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0145] Embodiment 1

[0146] This paper proposes a P2P electricity / heat energy interaction model for differentiated building prosumers, which optimizes the interaction process of prosumers to minimize their own energy costs throughout the day. The model decision variables include the P2P electricity / heat energy interaction power, the operating power of controllable resources such as energy storage batteries, heat pumps, and thermal storage equipment, and the power of energy purchase and sale from urban integrated energy system operators.

[0147] The objective function is as follows:

[0148]

[0149] in, represents the charging and discharging cost of the energy storage battery of building prosumer i during period t; represents the heat storage and release cost of the thermal storage equipment of building prosumer i during period t; It represents the cost of buying and selling electricity from building prosumer i to the operator of the urban integrated energy system during period t; represents the cost of purchasing heat energy from building prosumer i to the operator of the urban integrated energy system during period t; i represents the building prosumer number; t represents the time interval; T represents the interaction period; I represents the set of building prosumers; represents the charging power of the energy storage battery of building prosumer i during period t; represents the discharge power of the energy storage battery of building prosumer i during period t; represents the unit charging and discharging cost of the energy storage battery of building prosumer i during period t; represents the thermal storage power of the thermal storage equipment of building prosumer i during period t; represents the heat release power of the thermal storage equipment of building prosumer i during period t; represents the unit heat storage and release cost of the thermal storage equipment of building prosumer i during period t; and They represent the power of electricity and heat purchased by building prosumer i from the operator of the urban integrated energy system during period t, respectively. A positive value indicates electricity purchase, and a negative value indicates electricity sale; and They represent the prices of electricity and heat purchased by building prosumer i from the operator of the urban integrated energy system during period t.

[0150] There are six constraints in total, namely Constraint 1: P2P electricity / heat energy interaction constraint; Constraint 2: Energy purchase and sale constraints from urban integrated energy system operators; Constraint 3: Building thermal balance constraints; Constraint 4: Energy storage battery, heat storage equipment, radiator and heat pump operation constraints; Constraint 5: Building prosumer electricity / heat power balance constraints; Constraint 6: Distribution network constraints and heating network constraints.

[0151] Constraint 1: P2P electricity / heat interaction constraints:

[0152]

[0153]

[0154]

[0155]

[0156] in, and They represent the P2P electric energy and thermal energy interaction power between building prosumer i and building prosumer j during period t, respectively. A positive value indicates that energy is purchased from building prosumer j, and a negative value indicates that energy is sold to building prosumer j. and They represent the P2P electric energy and thermal energy interaction power between building prosumer j and building prosumer i during period t, respectively. A positive value indicates that energy is purchased from building prosumer i, and a negative value indicates that energy is sold to building prosumer i. and They represent the maximum values ​​of interactive electric energy sold and purchased by building prosumer i from building prosumer j respectively; and They represent the maximum interactive heat energy sold and purchased by building prosumer i to building prosumer j, respectively.

[0157] Constraint 2: Constraints on the purchase and sale of energy from urban integrated energy system operators:

[0158]

[0159]

[0160] in, and They represent the maximum power of electricity and heat purchased by building prosumer i from the operator of the urban integrated energy system.

[0161] Constraint 3: Building thermal balance constraints:

[0162] The present invention assumes that each floor of the building prosumer is composed of four heating zones. Taking heating zone 1 as an example, the RC network thermal dynamic model is established. Figure 1 First, by selecting a single wall RC network model of heating area 1, the meaning of thermal resistance and heat capacity parameters in the wall thermal dynamic process is explained, as shown in Figure 1 As shown in (a) in the figure. The two sides of the wall are the outdoor and indoor environments, which can be divided into indoor air nodes, wall nodes and outdoor air nodes. These nodes are connected to each other through thermal resistance and grounded through thermal capacitance. Their temperatures are T w 、T out and T r . C w , C out and C r are the heat capacities of the three nodes. Heat can be transferred into the room through two ways: through walls or windows. out is the thermal resistance in the heat transfer process between outdoor air and the outer surface of the wall, R r is the thermal resistance in the heat transfer process between indoor air and the inner surface of the wall, R wall is the thermal resistance of heat transfer inside the wall, and the sum of the three can be approximated as the total thermal resistance R of the wall. w . R win is the total thermal resistance of the window, Q rad is the heat gained by the wall from the sun, and the wall temperature T w Affected by wall thermal resistance, heat capacity and sunlight. rad,win The heat gained by the window from the sun. Indoor temperature T r The heat dissipation Q of indoor personnel and equipment is calculated by wall thermal resistance, heat capacity, window heat capacity, light intensity, and indoor personnel and equipment heat dissipation Q. int And the heat dissipation of temperature control load Q load etc. factors jointly determine the

[0163] by Figure 1 In (b), the indoor air temperature of heating area 1 is the reference node. Assume that this node is node 1 and its temperature is T 1 , whose value is equal to T1 r The air nodes of other areas adjacent to heating area 1 (heating area 4, heating area 2 and two outdoor areas) are nodes 2, 3, 4 and 5, and their temperatures are T 2 、T 3 、T 4 、T 5 The two outdoor areas are nodes 4 and 5, and the sunlight intensity they receive is and The temperatures of the four walls separating the two areas are The heat capacities are The thermal resistances are The present invention assumes that the temperature distribution in the heating area is uniform and the total indoor air mass is constant. The wall thermal dynamic model and the indoor thermal dynamic model in the heating area 1 are as follows:

[0164]

[0165]

[0166] Among them, r 1,2 、r 1,2 、r 1,2 、r 1,2 They are the states of sunlight exposure to the wall. If it is exposed to sunlight, the value is 1, otherwise it is 0; α 1,2 , α 1,3 , α 1,4 , α 1,5 are the heat absorption rates of the four walls, are the surface areas of the four walls respectively; is the indoor air heat capacity of heating zone 1; is the heat increment of the temperature control load in heating area 1; τ win is the window transmittance; A win is the window area. is the heat gain generated by indoor residents and equipment in heating zone 1, and its expression is shown in formula (10):

[0167] Q int =Q body +Q app (10);

[0168]

[0169] Q app =percent e ×(Light per +Appliance per )×S×ε (12);

[0170] Among them, the heat generated by human activities is Q body and the hourly indoor rate of people b 、Average heat dissipation of the human body per and the indoor area S. The calculation process is shown in formula (11); the random heat load Q when lighting appliances and equipment are working app Instead of using percentage hourly e , indoor area S, heat dissipation ratio ε, and power density of electrical appliances and equipment (Light per +Applianceper ), and its calculation process is shown in formula (12).

[0171] Using the building RC network model, the wall thermal balance and indoor thermal balance constraints of a single heating area 1 in building prosumer i are constructed. Since the model belongs to a differential equation, it is difficult to solve, so it is converted and organized into a difference equation:

[0172]

[0173]

[0174] In formula (14), the heat load demand of heating area 1 is met by a variety of heating equipment; formula (15) indicates that the heat load needs to be controlled within a certain range; formula (16) indicates that the intelligent regulating valve configured on the radiator can set the indoor temperature comfort range of users in the building to ensure the comfort of users.

[0175]

[0176]

[0177] in, and They are the upper and lower limits of heat load of heating zone 1 in building prosumer i during period t; is the indoor temperature of heating zone 1 in building prosumer i during period t; and are the upper and lower limits of the temperature comfort range respectively.

[0178] Constraint 4: Energy storage batteries, heat storage equipment, radiators and heat pumps operation constraints:

[0179] During the off-peak hours of building prosumers or when renewable energy sources such as solar photovoltaics are producing more electricity, the energy storage battery can be charged to store electricity; during the peak hours, the energy storage battery can be discharged to achieve a shift in electricity supply and demand. Its operating constraints include charging and discharging power constraints and SOC limits:

[0180]

[0181]

[0182]

[0183]

[0184]

[0185]

[0186] in, and They represent the charging and discharging power of the energy storage battery of building prosumer i during period t, respectively; and They represent the maximum charging and discharging power of the energy storage battery of building prosumer i during period t respectively; represents the SOC, i.e., the state of charge, of the energy storage battery of building prosumer i during period t; and They represent the upper and lower limits of the SOC of the energy storage battery of building prosumer i in period t respectively; represents the capacity of the energy storage battery of building prosumer i; and They represent the charging and discharging efficiency of the energy storage battery of building prosumer i respectively; represents the initial SOC of the energy storage battery of building prosumer i, indicating that the SOC of the energy storage battery remains unchanged at the initial period t0 of each day, and there are d periods every day; Indicates the SOC of the energy storage battery in time period t0+d; Represents the SOC of the energy storage battery in time period t-1.

[0187] Thermal storage equipment can store heat during periods of low electricity prices and release heat during periods of high electricity prices, thereby increasing the flexibility of heat production and use by building prosumers. The relationship between the heat storage capacity, heat storage power, heat release power and heat dissipation of thermal storage equipment is shown in the following formula:

[0188]

[0189] in, represents the heat storage of the heat storage equipment of building prosumer i in period t+1; represents the heat storage of the heat storage equipment of building prosumer i during period t; represents the thermal storage power of the thermal storage equipment of building prosumer i during period t; represents the heat release power of the heat storage equipment of building prosumer i during period t; η TES Indicates the heat dissipation rate of thermal storage equipment.

[0190] The following constraints need to be met when the thermal storage device is running:

[0191]

[0192]

[0193]

[0194]

[0195] in, and They represent the maximum heat storage and heat release power of the heat storage equipment of building prosumer i respectively; and They represent the minimum and maximum heat storage capacity of the thermal storage equipment of building prosumer i respectively.

[0196] In this embodiment, the hot water in the heat storage tank is only used for building heating and does not provide domestic water services for users. In addition, assuming that the upper and lower layers of hot water in the heat storage device are evenly mixed and the temperature of any point in each layer is equal, only the heat transfer process along the flow direction of hot water in the heat storage device needs to be considered.

[0197] Radiators can be applied in two modes: P2P electric energy interaction and P2P electric / heat energy interaction. When the flow of the radiator is adjustable, its thermal characteristics directly affect the temperature comfort of users in the building. At the same time, the flow regulation mode will also affect the formulation of the P2P energy interaction strategy of the building prosumer.

[0198] Building prosumers optimize the management of the thermal power delivered by the radiator according to the temperature comfort range. Assuming that the radiator is equipped with an intelligent regulating valve, the user can use it to adjust the water flow through the radiator, thereby changing its heat dissipation. Therefore, the thermal power delivered by the radiator in time period t is:

[0199]

[0200] Among them, c p represents the specific heat capacity of water; represents the water flow rate of building prosumer i in the radiator during period t; T s and T r Respectively represent the radiator water supply and return temperatures; It represents the thermal power delivered by the radiator to building prosumer i during period t. When the building prosumer adopts flow control, it is assumed that the supply and return water temperatures are constant and known.

[0201] The air source heat pump driven by electric energy uses a fan to exchange heat between outdoor air and the refrigerator in the evaporator to provide indoor heating. This embodiment uses a variable frequency air source heat pump (hereinafter referred to as a heat pump) to meet the heat demand of the building. Compared with the traditional 0-1 control method, the heat pump can extend the life of the compressor and reduce energy consumption by adjusting the compressor speed to avoid frequent start and stop. The power consumption of the heat pump is shown in the following formula:

[0202]

[0203] in, represents the power consumption of the heat pump of building prosumer i during period t; represents the heating power of the heat pump of building prosumer i during period t; COP HP Indicates the energy efficiency ratio of the heat pump.

[0204] During operation, the heat pump must meet the following constraints:

[0205]

[0206] in, represents the maximum power consumption of the heat pump of building prosumer i; Represents the power consumption of the heat pump of building prosumer i during period t.

[0207] Constraint 5. Electricity / heat power balance constraints for building prosumers:

[0208]

[0209]

[0210]

[0211] in, represents the total heat load of building prosumer i during period t; represents the power of the rigid electrical load of building prosumer i during period t; Indicates the actual photovoltaic power generation output; represents the actual photovoltaic power generation output of building prosumer i during period t; j represents the building prosumer number; represents the power of heat energy produced by the heat pump of building prosumer i during period t; represents the heat load of heating zone n in building prosumer i during period t; n represents the number of heating zones; N represents the total number of heating zones;

[0212] Constraint 6. Distribution network constraints and heating network constraints:

[0213] It is worth noting that the operator of the urban integrated energy system needs to interact with the building prosumers under the P2P electricity / heat energy interaction to ensure that its interaction scheme meets the operation constraints of the distribution network and the heating network. For the distribution network constraints, the linearized DistFlow model is used. Figure 2 One of the branches is shown in the figure, with node m as the starting point and node m+1 as the end point, and power flows from node m to node m+1. The active power, reactive power and voltage of this branch can be expressed as:

[0214]

[0215]

[0216]

[0217] 1-ε≤V m ≤1+ε(37);

[0218] Among them, P m+1 represents the active power of node m+1; P m represents the active power of node m; r f Indicates line resistance; Q m represents the reactive power of node m; represents the active power of the load at node m+1; Q m+1 represents the reactive power of node m+1; x f Indicates line reactance; represents the reactive power of the load at node m+1; P m +iQ m represents the complex power injected by node m; V m Represents the voltage of node m; V m+1 represents the voltage of node m+1; represents the complex power of the load at node m+1; r f +ix f represents line impedance; V0 represents the branch root node voltage; ε represents the node voltage operating fluctuation range;

[0219] The heating network model is as follows:

[0220] Y HDN m pipe =m node (38);

[0221]

[0222]

[0223]

[0224]

[0225] Among them, Y HDN Represents the incidence matrix of the heating network; m pipe and m node Respectively represent each pipeline flow and node flow vector; and They represent the maximum and minimum water flow in pipeline l during period t respectively; N pipe Indicates the number of pipelines; p n,t represents the water pressure at node n during period t; ξ l represents the pipeline pressure coefficient; κ l , L l and d l They represent the friction coefficient, length and inner diameter of the pipe l respectively; ρ water Indicates water density; N node Indicates the number of nodes; and Respectively represent the maximum and minimum values ​​of the node pressure; represents the hot water flow in pipe l during period t; p n+1,t represents the water pressure at node n+1 during period t;

[0226] Formula (38) indicates that the flow rate injected into the node in the heating network is equal to the flow rate out of the node. At the same time, due to the short length and small scale of the heating network, the present invention ignores the heat loss in the heating network; Formula (39) indicates that the water flow rate in the pipeline needs to be constrained within a certain range to prevent pipeline vibration; Formula (40) indicates that the pipeline pressure drop caused by the friction of water flow along the pipeline is proportional to the square of the water flow rate; Formula (42) indicates that the node pressure needs to be controlled within a certain range to ensure the safety of the heating network.

[0227] For the heating network constraints, the present invention will adopt the above heating network model, where the hydraulic constraint (40) contains the square term of the variable Therefore, the piecewise linearization method is used to deal with it. By constraining (39) It is divided into Q (q∈Q) segments (with a value of 100) and represented on the abscissa. The corresponding ordinate and slope are shown in the following formula:

[0228]

[0229] Then the nonlinear equation in formula (40) is processed piecewise and replaced by the sum of a set of linear equations, as shown in formula (44)-formula (46):

[0230]

[0231]

[0232]

[0233] In the formula, is a 0-1 variable to represent Whether the value is on the q segment.

[0234] The above objective functions and constraints constitute the differentiated building prosumer P2P electricity / heat energy interaction model of the present invention.

[0235] S2. An asynchronous distributed interactive solution algorithm based on ADMM is used to solve the P2P electricity / heat energy interaction model of differentiated building prosumers, including the following sub-steps:

[0236] By using the ADMM algorithm and decomposing the coupling variables, multiple differentiated building prosumer P2P interaction sub-problems can be obtained. On this basis, an asynchronous distributed solution algorithm integrating the prediction step and the anomaly detection step is used to solve the decomposed building prosumer P2P electricity / heat energy interaction sub-problems in an asynchronous manner.

[0237] Based on the ADMM algorithm, the P2P electricity / heat energy interaction model can be decomposed into multiple building prosumer interaction sub-problems. Each building prosumer can solve its sub-problem independently and share limited exchange information with other building prosumers. The key to decoupling is to determine the coupling variables in the building prosumer sub-problem, and then introduce auxiliary variables to solve the coupling variables. To ensure that the coupling variables are equal, a consistency constraint can be formed between sub-problems:

[0238]

[0239]

[0240] To this end, the Lagrangian relaxation method is used to extend the above consistency constraints into the objective function (1):

[0241]

[0242] Among them, χ i Represents a local variable; x i represents the coupling variable, i.e., the P2P electricity / heat interaction power of building prosumer i; z i(j) Represents auxiliary variables; and They represent the P2P electric energy and thermal energy interaction power between building prosumer i and building prosumer j in the k+1th iteration period t; Cost i The same variables as in the objective function (1), Both are used to represent the P2P electric / thermal energy exchange coupling between differentiated building prosumers; ρ represents the step size; It represents the Lagrange multiplier corresponding to the consistency constraint in time period t, that is, the P2P electricity / heat energy interaction price.

[0243] Through the above decoupling, if the synchronous ADMM algorithm is used to solve the sub-problems of differentiated building prosumers, it will cause the building prosumers with better computing and communication facilities to wait for other building prosumers with poorer performance to share their coupling variables after solving their own sub-problems, so as to update the auxiliary variable values ​​and Lagrange multipliers in each interaction iteration, such as Figure 3 As shown, this will affect the computational efficiency of solving interactive problems.

[0244] like Figure 3As shown, the present invention adopts an asynchronous distributed interactive solution algorithm based on ADMM to solve the above challenges, and the specific implementation is as follows:

[0245] (1) Prediction step

[0246] Assume that building prosumer i is in the kth i In the kth iteration, if the coupling variable value transmitted by building prosumer p is not received, building prosumer i will predict the value of the variable. i -2nd and kth i -1 The coupling variable values ​​in the iteration are predicted using momentum extrapolation:

[0247]

[0248]

[0249] in, Indicates the kth i -2nd and kth i -1 The trend of the coupling variable value in the iteration; μ and η are weight factors; Indicates the kth i -3rd and kth i - Trend of coupling variable values ​​in 2 iterations; and Respectively represent the kth i -1 and k i - The coupling variables of building prosumer i and building prosumer p in the 2nd iteration; Indicates the kth i The predicted value of the coupled variable between building prosumer i and building prosumer p in the iteration;

[0250] It is worth noting that k in formula (50) and formula (51) i Need to be greater than 2; when k i When it is less than or equal to 2, the building prosumer will directly adopt the coupling variable value of the first iteration or the initial setting.

[0251] The predicted value of the coupling variable obtained based on formula (50) and formula (51) and the actual value of the coupled variable The kth i +1 Auxiliary variable values ​​and Lagrange multipliers for iteration:

[0252]

[0253]

[0254]

[0255]

[0256] In the formula, Indicates the kth i +1 iteration of the auxiliary variable of electric energy for building prosumer i in period t; Indicates the kth i The thermal energy auxiliary variable of building prosumer i in the iteration t period; and Respectively represent the kth i The P2P electric and thermal energy interaction power between building prosumer i and building prosumer a during the iteration t period; and Respectively represent the kth i The P2P electric and thermal energy interaction power between building prosumer i and building prosumer p during the iteration t period; and Respectively represent the kth i The Lagrange multiplier of the electrical and thermal energy of building prosumer i during the iteration t period; and Respectively represent the kth i -1 Lagrangian multipliers of electrical and thermal energy of building prosumer i and building prosumer a during period t of iteration t; and Respectively represent the kth i -1 Lagrange multipliers of electrical and thermal energy of building prosumer i and building prosumer p in time period t; and Respectively represent the kth i The auxiliary variables of electric energy and thermal energy of building prosumer i and building prosumer a in the iteration t period; and Respectively represent the kth i The auxiliary variables of electrical energy and thermal energy of building prosumer i and building prosumer p in the iteration t period;

[0257] To this end, we can construct the building prosumer i in the kth i The sub-problem of +1 iteration is as follows:

[0258]

[0259] Among them, N a N represents the total number of building prosumer sellers; p Represents the total number of building prosumer buyers.

[0260] (2) Anomaly detection step.

[0261] Based on the energy costs of building prosumers obtained in multiple iterations, the predicted values ​​of the coupling variables are monitored and tested, and it is evaluated whether the predicted values ​​need to be adjusted. Prediction errors may affect the Lagrange multipliers updated based on the predicted values, resulting in the solved interaction scheme being a suboptimal solution.

[0262] First, based on the energy cost of the building prosumer i in the previous s iterations, the corresponding cost curve is obtained. Then, the wavelet threshold method and moving average method are used to reduce the noise of the curve until the noise-reduced curve and the original curve f i The difference between them is less than the threshold ε, as shown in formula (57) and formula (58):

[0263]

[0264]

[0265] in, All represent the Lagrange multipliers corresponding to the consistency constraints;

[0266] Subsequently, based on the above difference, anomaly detection is performed on the predicted value in the k+1th iteration. Specifically, the difference between the denoised curve and the original curve result of the sth iteration is divided into ψ groups using the K-Means method, and then the group that will cause anomalies is determined. i When the difference of an iteration belongs to a group that will cause an exception, the Lagrange multiplier of that iteration will not be updated, thus ensuring that the above asynchronous distributed solution algorithm generates the optimal interaction solution.

[0267] Figure 4 The solution process of asynchronous distributed interaction between building prosumers is given. The specific steps are as follows:

[0268] Step 1: Construct a differentiated building prosumer P2P electricity / heat energy interaction model and set the index k=1 of each building prosumer.

[0269] Step 2: Set the initial values ​​of prosumer coupling variables, auxiliary variables, Lagrange multipliers, and step sizes for each building.

[0270] Step 3: Decouple the original model in step 1 into multiple sub-problems containing Lagrangian functions and internal constraints of building prosumers as described in formula (56).

[0271] Step 4: Each building prosumer uses the latest auxiliary variable values ​​and Lagrange multipliers to solve its subproblem in an asynchronous manner.

[0272] Step 5: Perform the prediction step and use Formula (52)-Formula (55) to update the auxiliary variable values ​​and Lagrange multipliers for the kth iteration.

[0273] Step 6: Perform the anomaly detection step and evaluate the prediction result of the kth iteration. If the predicted value will cause an exception, only update the auxiliary variable value and set k=k+1, then return to step 4.

[0274] Step 7: If all the primal residuals and dual residuals are less than the convergence threshold, as shown in Formula (59)-Formula (60), the iteration stops; otherwise, return to Step 4.

[0275]

[0276]

[0277] Among them, r primal represents the original residual; Indicates the kth i The coupling variables of the iteration; Indicates the kth i Auxiliary variable for the iteration; r dual represents the dual residual; ρ represents the step size; Indicates the kth i -1 Auxiliary variables for iteration; δ1 and δ2 both represent convergence thresholds.

[0278] Embodiment 2

[0279] like Figure 5 As shown, the present invention also provides a differentiated building prosumer P2P electricity / heat energy interactive operation device, including a differentiated building prosumer P2P electricity / heat energy interactive model construction module and a differentiated building prosumer P2P electricity / heat energy interactive model solving module; the differentiated building prosumer P2P electricity / heat energy interactive model solving module includes five submodules, namely, an initialization module, a sub-problem solving module, a prediction module, an anomaly detection module and a discrimination module;

[0280] Among them, the differentiated building prosumer P2P electricity / heat energy interaction model construction module is used to construct the differentiated building prosumer P2P electricity / heat energy interaction model;

[0281] The differentiated building prosumer P2P electricity / heat energy interaction model solving module is used to solve the differentiated building prosumer P2P electricity / heat energy interaction model through an asynchronous distributed interactive solving algorithm based on ADMM;

[0282] The initialization module is used to set the index of each building prosumer and the initial values ​​of the coupling variables, auxiliary variables, Lagrange multipliers and step length of each building prosumer. The ADMM algorithm is used to decompose the coupling variables to obtain the differentiated building prosumer P2P electricity / heat interaction sub-problems;

[0283] The sub-problem solving module is used to solve the sub-problems of each building prosumer in an asynchronous manner using the latest auxiliary variable values ​​and Lagrange multipliers;

[0284] The prediction module is used to perform the prediction step and update the auxiliary variable values ​​and Lagrange multipliers of the kth iteration;

[0285] The anomaly detection module is used to perform the anomaly detection step and evaluate the prediction results of the kth iteration;

[0286] The discriminant module is used to judge the original residual, dual residual and convergence threshold.

[0287] Therefore, the present invention adopts the above-mentioned differentiated building prosumer P2P electricity / heat energy interactive operation method and device. This interactive method can significantly reduce the energy consumption cost of each building prosumer, improve the solution efficiency, and ensure the global optimal solution.

[0288] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A differentiated building prosumer P2P electricity / heat energy interactive operation method, characterized in that: The following steps are involved: S1. Construct a P2P electricity / heat energy interaction model for differentiated building prosumers; The objective function of the P2P electricity / heat energy interaction model of differentiated building prosumers is as follows: in, represents the charging and discharging cost of the energy storage battery of building prosumer i during period t; represents the heat storage and release cost of the thermal storage equipment of building prosumer i during period t; It represents the cost of buying and selling electricity from building prosumer i to the operator of the urban integrated energy system during period t; represents the cost of purchasing heat energy from building prosumer i to the operator of the urban integrated energy system during period t; i represents the building prosumer number; t represents the time interval; T represents the interaction period; I represents the set of building prosumers; represents the charging power of the energy storage battery of building prosumer i during period t; represents the discharge power of the energy storage battery of building prosumer i during period t; represents the unit charging and discharging cost of the energy storage battery of building prosumer i during period t; represents the thermal storage power of the thermal storage equipment of building prosumer i during period t; represents the heat release power of the thermal storage equipment of building prosumer i during period t; represents the unit heat storage and release cost of the thermal storage equipment of building prosumer i during period t; and They represent the power of electricity and heat purchased by building prosumer i from the operator of the urban integrated energy system during period t, respectively. A positive value indicates electricity purchase, and a negative value indicates electricity sale; and They represent the prices of electricity and heat purchased by building prosumer i from the operator of the urban integrated energy system during period t; The constraints of the P2P electricity / heat energy interaction model for differentiated building prosumers include the following six types: Constraint 1: P2P electricity / heat interaction constraint; Constraint 2: Constraints on the purchase and sale of energy from urban integrated energy system operators; Constraint 3: Building thermal balance constraint; Constraint 4: Energy storage battery, heat storage equipment, radiator and heat pump operation constraints; Constraint 5: Electricity / heat power balance constraint of building prosumers; Constraint 6: Distribution network constraints and heating network constraints; S2. An asynchronous distributed interactive solution algorithm based on ADMM is used to solve the P2P electricity / heat energy interaction model of differentiated building prosumers, including the following sub-steps: S21, setting the index k of each building prosumer to 1, and setting the initial values ​​of the coupling variables, auxiliary variables, Lagrange multipliers and step length of each building prosumer, using the ADMM algorithm, and decomposing the coupling variables to obtain the differentiated building prosumer P2P electricity / heat energy interaction sub-problem; S22, each building prosumer uses the latest auxiliary variable values ​​and Lagrange multipliers to solve its subproblem in an asynchronous manner; S23, executing the prediction step, updating the auxiliary variable value and Lagrange multiplier of the kth iteration; S24, performing anomaly detection steps to evaluate the prediction results of the kth iteration; When the prediction result does not cause an exception, proceed to step S25; When the prediction result causes an exception, only the auxiliary variable value is updated, and k=k+1 is set, and then the process returns to step S22; S25, when all the original residuals and the dual residuals are less than the convergence threshold, as shown in formula (2) and formula (3), the iteration stops; otherwise, return to step S22; Among them, r primal represents the original residual; Indicates the kth i The coupling variables of the iteration; Indicates the kth i Auxiliary variable for the iteration; r dual represents the dual residual; ρ represents the step size; Indicates the kth i -1 Auxiliary variables for iteration; δ1 and δ2 both represent convergence thresholds.

2. A differentiated building prosumer P2P electricity / heat energy interactive operation method according to claim 1, characterized in that: The P2P electrical / thermal energy interaction constraints are as follows: in, and They represent the P2P electric energy and thermal energy interaction power between building prosumer i and building prosumer j during period t, respectively. A positive value indicates that energy is purchased from building prosumer j, and a negative value indicates that energy is sold to building prosumer j. and They represent the P2P electric energy and thermal energy interaction power between building prosumer j and building prosumer i during period t, respectively. A positive value indicates that energy is purchased from building prosumer i, and a negative value indicates that energy is sold to building prosumer i. and They represent the maximum values ​​of interactive electric energy sold and purchased by building prosumer i from building prosumer j respectively; and They represent the maximum interactive heat energy sold and purchased by building prosumer i to building prosumer j, respectively.

3. A differentiated building prosumer P2P electricity / heat energy interactive operation method according to claim 2, characterized in that: The constraints on the purchase and sale of energy from the city's integrated energy system operator are as follows: Among them, P i Grid,max and They represent the maximum power of electricity and heat purchased by building prosumer i from the operator of the urban integrated energy system.

4. A differentiated building prosumer P2P electricity / heat energy interactive operation method according to claim 3, characterized in that: The building heat balance constraints are as follows: Among them, 1 represents the reference node of the heating area, recorded as node 1; 2 and 3 represent the nodes of the two areas adjacent to the heating area, recorded as nodes 2 and 3; 4 and 5 represent the nodes of the two outdoor areas, recorded as nodes 4 and 5; w represents the node number; there is a window installed on the wall of node 5; the heat capacities of the four walls around the heating area of ​​building prosumer i are The thermal resistances are The wall temperatures during period t are The wall temperatures during the t+1 period are represents the temperature of node 1 of building prosumer i during period t, They represent the temperature of node 2, node 3, node 4, and node 5 of building prosumer i in period t respectively; r 1,2 、r 1,3 、r 1,4 、r 1,5 They represent the state of sunlight exposure to the wall. If it is exposed to sunlight, the value is 1, otherwise it is 0; α 1,2 , α 1,3 , α 1,4 , α 1,5 Respectively represent the heat absorption rate of the four walls; Respectively represent the surface areas of the four walls; They represent the solar radiation intensity received by the four walls of building prosumer i respectively; Δt represents the time difference; represents the indoor air heat capacity of node 1 of building prosumer i; T i out represents the outdoor temperature of building prosumer i; represents the indoor temperature of building prosumer i during period t; represents the indoor temperature of building prosumer i in period t+1; represents the temperature of the wall between the building prosumer i and w nodes during period t; represents the thermal resistance of windows of building prosumer i; represents the heat increment of the temperature control load in node 1 of building prosumer i during period t; represents the heat gain generated by indoor residents and equipment in node 1 of building prosumer i during period t; τ win represents the transmittance of the window; represents the area of ​​the windows of building prosumer i; It represents the solar radiation intensity of the windows of building prosumer i during period t; in, represents the heat load of building prosumer i at node 1 during period t; and They represent the upper and lower limits of the heat load of node 1 in building prosumer i during period t respectively; represents the indoor temperature of node 1 in building prosumer i during period t; T t r,max and T t r,min They represent the upper and lower limits of the temperature comfort range during period t respectively.

5. A differentiated building prosumer P2P electricity / heat energy interactive operation method according to claim 4, characterized in that: Energy storage battery operation constraints include charge and discharge power constraints and SOC limit constraints: in, and They represent the charging and discharging power of the energy storage battery of building prosumer i during period t; P i ba ,chg,max and P i ba,dischg,max They represent the maximum charging and discharging power of the energy storage battery of building prosumer i during period t respectively; represents the SOC, i.e., the state of charge, of the energy storage battery of building prosumer i during period t; and They represent the upper and lower limits of the SOC of the energy storage battery of building prosumer i in period t respectively; represents the capacity of the energy storage battery of building prosumer i; and They represent the charging and discharging efficiency of the energy storage battery of building prosumer i respectively; represents the initial SOC of the energy storage battery of building prosumer i, indicating that the SOC of the energy storage battery remains unchanged at the initial period t0 of each day, and there are d periods every day; Indicates the SOC of the energy storage battery in time period t0+d; Represents the SOC of the energy storage battery in time period t-1; The relationship between the heat storage capacity, heat storage power, heat release power and heat dissipation of the heat storage equipment is as follows: in, represents the heat storage of the heat storage equipment of building prosumer i in period t+1; represents the heat storage of the heat storage equipment of building prosumer i during period t; represents the thermal storage power of the thermal storage equipment of building prosumer i during period t; represents the heat release power of the heat storage equipment of building prosumer i during period t; η TES Indicates the heat dissipation rate of the thermal storage device; The operating constraints of thermal storage equipment are as follows: in, and They represent the maximum heat storage and heat release power of the heat storage equipment of building prosumer i; W i TES,min and W i TES,max They represent the minimum and maximum heat storage capacity of the thermal storage equipment of building prosumer i respectively; The heat sink operation constraints are as follows: Among them, c p represents the specific heat capacity of water; represents the water flow rate of building prosumer i in the radiator during period t; T s and T r Respectively represent the radiator water supply and return temperatures; represents the heat power transferred by the radiator to the prosumer i of the building during period t; The power consumption of heat pump is as follows: in, represents the power consumption of the heat pump of building prosumer i during period t; represents the heating power of the heat pump of building prosumer i during period t; COP HP It represents the energy efficiency ratio of the heat pump; The heat pump operation constraints are as follows: Among them, P i HP,max represents the maximum power consumption of the heat pump of building prosumer i; Represents the power consumption of the heat pump of building prosumer i during period t.

6. A differentiated building prosumer P2P electricity / heat energy interactive operation method according to claim 5, characterized in that: The electrical / thermal power balance constraints of building prosumers are as follows: in, represents the total heat load of building prosumer i during period t; represents the power of the rigid electrical load of building prosumer i during period t; Indicates the actual photovoltaic power generation output; represents the actual photovoltaic power generation output of building prosumer i during period t; j represents the building prosumer number; represents the power of heat energy produced by the heat pump of building prosumer i during period t; represents the heat load of heating zone n in building prosumer i during period t; n represents the number of heating zones; N represents the total number of heating zones; The distribution network constraints use the linearized DistFlow model. The active power, reactive power and voltage of any branch are expressed as: 1-ε≤V m ≤1+ε(34); Among them, P m+1 represents the active power of node m+1; P m represents the active power of node m; r f Indicates line resistance; Q m represents the reactive power of node m; represents the active power of the load at node m+1; Q m+1 represents the reactive power of node m+1; x f Indicates line reactance; represents the reactive power of the load at node m+1; P m +iQ m represents the complex power injected by node m; V m Represents the voltage of node m; V m+1 represents the voltage of node m+1; represents the complex power of the load at node m+1; r f +ix f represents line impedance; V0 represents the branch root node voltage; ε represents the node voltage operating fluctuation range; The heating network constraint adopts the heating network model, which is as follows: Y HDN m pipe =m node (35); Among them, Y HDN Represents the incidence matrix of the heating network; m pipe and m node Respectively represent each pipeline flow and node flow vector; and They represent the maximum and minimum water flow in pipeline l during period t respectively; N pipe Indicates the number of pipelines; p n,t represents the water pressure at node n during period t; ξ l represents the pipeline pressure coefficient; κ l , L l and d l They represent the friction coefficient, length and inner diameter of the pipe l respectively; ρ water Indicates water density; N node Indicates the number of nodes; and Respectively represent the maximum and minimum values ​​of the node pressure; represents the hot water flow in pipe l during period t; p n+1,t represents the water pressure at node n+1 during period t; Among them, formula (37) contains the square term of the variable Therefore, the piecewise linearization method is used to deal with it; by reducing the range of formula (36) Divide into Q segments, q∈Q and represent them on the abscissa, and the corresponding ordinate and slope are as follows: Then the nonlinear equation in formula (37) is processed piecewise and replaced by the sum of a set of linear equations, as shown in formula (41)-formula (43): in, is a 0-1 variable, indicating Whether the value is on the q segment.

7. A differentiated building prosumer P2P electricity / heat energy interactive operation method according to claim 6, characterized in that: In step S22, each building prosumer uses the latest auxiliary variable value and Lagrange multiplier to solve its sub-problem in an asynchronous manner. The specific operation is: Determine the coupled variables in the building prosumer subproblem and then introduce auxiliary variables To ensure that the coupling variables are equal, thus forming consistency constraints between sub-problems: The Lagrangian relaxation method is used to extend the above consistency constraint into the objective function (1): Among them, χ i Represents a local variable; x i represents the coupling variable, i.e., the P2P electricity / heat interaction power of building prosumer i; z i(j) Represents auxiliary variables; and They represent the P2P electric energy and thermal energy interaction power between building prosumer i and building prosumer j in the k+1th iteration t period; Cost i The same variables as in the objective function (1), Both are used to represent the P2P electric / thermal energy exchange coupling between differentiated building prosumers; ρ represents the step size; It represents the Lagrange multiplier corresponding to the consistency constraint in time period t, that is, the P2P electricity / heat energy interaction price.

8. A differentiated building prosumer P2P electricity / heat energy interactive operation method according to claim 7, characterized in that: In step S23, the prediction step is executed to update the auxiliary variable value and the Lagrange multiplier of the kth iteration. The specific operations are as follows: Assume that building prosumer i is in the kth i In the kth iteration, if the coupling variable value transmitted by building prosumer p is not received, building prosumer i will predict the value of the variable and use the kth iteration to predict the value of the variable. i -2nd and kth i -1 The coupling variable values ​​in the iteration are predicted using momentum extrapolation: in, Indicates the kth i -2nd and kth i -1 The trend of the coupling variable value in the iteration; μ and η are weight factors; Indicates the kth i -3rd and kth i - Trend of coupling variable values ​​in 2 iterations; and Respectively represent the kth i -1 and k i - The coupling variables of building prosumer i and building prosumer p in the 2nd iteration; Indicates the kth i The predicted value of the coupled variable between building prosumer i and building prosumer p in the iteration; The predicted value of the coupling variable obtained based on formula (47) and formula (48) and the actual value of the coupled variable Calculate the kth i +1 Auxiliary variable values ​​and Lagrange multipliers for iteration: In the formula, Indicates the kth i +1 iteration of the auxiliary variable of electric energy for building prosumer i in period t; Indicates the kth i The thermal energy auxiliary variable of building prosumer i in the iteration t period; and Respectively represent the kth i The P2P electric and thermal energy interaction power between building prosumer i and building prosumer a during the iteration t period; and Respectively represent the kth i The P2P electric and thermal energy interaction power between building prosumer i and building prosumer p during the iteration t period; and Respectively represent the kth i The Lagrange multiplier of the electrical and thermal energy of building prosumer i during the iteration t period; and Respectively represent the kth i -1 Lagrangian multipliers of electrical and thermal energy of building prosumer i and building prosumer a during period t of iteration t; and Respectively represent the kth i -1 Lagrange multipliers of electrical and thermal energy of building prosumer i and building prosumer p in time period t; and Respectively represent the kth i The auxiliary variables of electric energy and thermal energy of building prosumer i and building prosumer a in the iteration t period; and Respectively represent the kth i The auxiliary variables of electrical energy and thermal energy of building prosumer i and building prosumer p in the iteration t period; Construct the building prosumer i in the kth i The sub-problem of +1 iteration is as follows: Where N a N represents the total number of building prosumer sellers; p Represents the total number of building prosumer buyers.

9. A differentiated building prosumer P2P electricity / heat energy interactive operation method according to claim 8, characterized in that: In step S24, the abnormality detection steps are as follows: Based on the energy cost of the building prosumer i in the previous s iterations, the corresponding cost curve is obtained; The wavelet threshold method and moving average method are used to reduce the noise of the cost curve until the noise-reduced curve f i Denoise and the original curve f i The difference between the original curve f i The ratio of is less than the threshold ε, as shown in formula (54) and formula (55): in, All represent the Lagrange multipliers corresponding to the consistency constraints; Based on the above difference, anomaly detection is performed on the predicted value in the k+1th iteration.

10. A device for implementing the differentiated building prosumer P2P electricity / heat energy interactive operation method according to any one of claims 1 to 9, characterized in that: It includes a differentiated building prosumer P2P electricity / heat energy interaction model construction module and a differentiated building prosumer P2P electricity / heat energy interaction model solution module; The differentiated building prosumer P2P electricity / heat interaction model solution module includes five submodules, namely, initialization module, subproblem solution module, prediction module, anomaly detection module and discrimination module; Among them, the differentiated building prosumer P2P electricity / heat energy interaction model construction module is used to construct the differentiated building prosumer P2P electricity / heat energy interaction model; The differentiated building prosumer P2P electricity / heat energy interaction model solving module is used to solve the differentiated building prosumer P2P electricity / heat energy interaction model through an asynchronous distributed interactive solving algorithm based on ADMM; The initialization module is used to set the index of each building prosumer and the initial values ​​of the coupling variables, auxiliary variables, Lagrange multipliers and step length of each building prosumer. The ADMM algorithm is used to decompose the coupling variables to obtain the differentiated building prosumer P2P electricity / heat interaction sub-problems; The sub-problem solving module is used to solve the sub-problems of each building prosumer in an asynchronous manner using the latest auxiliary variable values ​​and Lagrange multipliers; The prediction module is used to perform the prediction step and update the auxiliary variable values ​​and Lagrange multipliers of the kth iteration; The anomaly detection module is used to perform the anomaly detection step and evaluate the prediction results of the kth iteration; The discriminant module is used to judge the original residual, dual residual and convergence threshold.

Images