A method for optimizing the scheduling of multi-agent integrated energy systems under information asynchrony

Abstract

The present invention belongs to the field of smart grids, and discloses a method for optimizing and dispatching a multi-agent integrated energy system under information asynchrony. The method establishes a mathematical model for the joint operation of multi-agent IES, and adopts an asynchronous ADMM algorithm to solve the information asynchrony problem existing in the solution process of IES belonging to different subjects, thereby effectively handling the information asynchrony problem. When a power shortage occurs in the system, the method of the present invention adopts an asynchronous ADMM algorithm to solve the communication delay problem caused by the asynchrony of data sampling in the mutual coordination of heat / electricity of each IES in a distributed manner. The computing nodes can be updated at different speeds and time points according to their own conditions, without waiting for other nodes to synchronize, which greatly improves the flexibility and adaptability of the calculation, effectively handles the information asynchrony problem, and improves the calculation efficiency of the algorithm.

Description

Technical Field

[0001] The present invention belongs to the field of smart grids, and in particular relates to a method for optimizing and dispatching a multi-agent integrated energy system under the condition of information asynchrony. Background Art

[0002] The rapid growth of electricity demand has brought new challenges to the supply and regulation of distribution networks. With the widespread access of distributed power generation units, distribution networks need to strengthen the regulation of the demand side to maintain stable operation of the system. The interconnection and mutual assistance of demand-side resources can realize the deployment of flexible loads in a larger time and space range, and improve the flexibility of the distribution network system. In the new distribution network system, due to the access to multiple energy sources such as wind power and photovoltaics, especially the access to the integrated energy system (IES), the demand side integrates multiple loads such as electricity and heat. When power shortage occurs, the distribution network needs to regulate the electric load and thermal load to improve the utilization rate of renewable energy. The study pointed out that by tapping the dispatchable potential of flexible resources on the demand side, the regulation effect of flexible loads can be achieved.

[0003] While IES interconnection realizes energy complementarity, since different IES belong to different stakeholders, when multiple subject IES are connected to the active distribution network, in order to balance the interests of different operating subjects, reduce computational complexity, and improve the privacy of user information, a distributed optimization algorithm based on the Alternating Direction Multiplier Method (ADMM) is used to solve this type of multi-subject power allocation problem. However, the traditional ADMM algorithm solves the energy coordination strategy between multi-subject IES under ideal conditions, that is, the algorithm can only be updated and solved after all the interactive information of each subject reaches the computing center, and all computing nodes operate at the same rhythm. Each round of iteration requires global synchronous communication to ensure the consistency of each variable value. However, in actual operation, due to the different monitoring and collection of data by IES of different subjects, synchronous update of parameters may cause communication delay or packet loss, thereby affecting the efficiency of solving the scheduling strategy. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides an optimization scheduling method for a multi-agent integrated energy system under the condition of information asynchrony. For the communication delay problem caused by the data sampling asynchrony in the thermal / electric coordination of each IES, an asynchronous ADMM algorithm is adopted for distributed solution to solve the information asynchrony problem existing in the coordination between multi-agent IES when a power shortage occurs in the system. At the same time, the stable operation of the system is achieved through the thermal / electric power coordination between the IES.

[0005] The present invention provides a method for optimizing and dispatching a multi-agent integrated energy system under information asynchrony, comprising the following steps:

[0006] Step 1: Construct a multi-agent IES hierarchical distributed coordination control framework;

[0007] Step 2: Establish the IES operation objective function with the goal of minimizing the CHP unit operation cost, the cost of interacting with the external grid, the operation and maintenance cost of the energy storage system, and the load shedding cost;

[0008] Step 3: Based on the IES operation objective function, establish the IES supply and demand balance and system operation constraints;

[0009] Step 4: construct an electric / thermal cooperative mutual assistance model between multiple IES entities. When a power shortage occurs, the electric / thermal cooperative mutual assistance model between multiple IES entities is used to perform cooperative optimization, thereby achieving system balance and ensuring stable operation and efficiency optimization of the system.

[0010] Step 5: Based on the electric / thermal cooperative mutual assistance model among the multi-agent IES, an asynchronous ADMM algorithm is used to solve the problem, realize the optimal scheduling strategy, and solve the problem of information asynchrony.

[0011] Furthermore, step 2 is specifically as follows:

[0012] Assuming there are N IES, the operating cost of IES-i is as shown in formula (1), where 1≤i≤N:

[0013] (1)

[0014] In the formula, is the total cost of IES-i when running alone, is the CHP unit operating cost, The cost of interacting with the external network is is the operation and maintenance cost of the energy storage system, Costs for users in IES-i to participate in shedding interruptible loads;

[0015] (1) The operating cost of a CHP unit is:

[0016] (2)

[0017] Where a1, b1, c1 are the CHP unit operating cost coefficients. is the total output power of the CHP unit, t is a certain moment when the CHP unit is running, and T is the total operating time of the CHP unit;

[0018] (2) Interaction costs with the external grid, including natural gas transaction costs and electricity transaction costs:

[0019] (3)

[0020] In the formula, , is the buy / sell gas price, , Indicates the amount of gas purchased and sold. , is the buying / selling price of electricity, , Indicates the electric power purchased from the power grid and the electric power sold to the large power grid;

[0021] (3) Considering that the service life of energy storage is limited, a certain amount of loss will occur during operation. The cost of the loss is expressed as:

[0022] (4)

[0023] In the formula , They represent the charging and discharging maintenance coefficients of the energy storage system in IES-i respectively; , Respectively represent the charging and discharging power of the energy storage system;

[0024] (4) In order to maintain the safety of the system, when IES has a power shortage, it will interrupt part of the flexible load to ensure the balance between output and load. The cost paid to the interruptible load is:

[0025] (5)

[0026] In the formula, , They are the compensation unit prices for cutting off interruptible load and thermal load in IES-i respectively; , They represent the interruptible electrical load and thermal load power removed in IES-i respectively.

[0027] Furthermore, step 3 is specifically as follows:

[0028] (1) Electric power balance:

[0029] (6)

[0030] In the formula, , are the output power of photovoltaic and wind power respectively, The power supplied to the CHP unit for the electrical load, , Indicates the electric power consumed by CCS and P2G, Indicates the original electrical load demand;

[0031] (2) Thermal power balance:

[0032] (7)

[0033] In the formula, , Indicates the thermal power output of the CHP unit and gas boiler GB. Indicates the original thermal power load;

[0034] (3) Gas balance constraints:

[0035] (8)

[0036] In the formula, is the natural gas consumption of the CHP unit, is the natural gas consumption in GB, The amount of natural gas produced for P2G;

[0037] (4) System safety operation constraints:

[0038] The normal operation of the system includes the combined heat and power unit CHP, the carbon capture device CCS, and the power-to-gas device P2G, and the modeling is as follows:

[0039] 1) The heat and power coupling output model of cogeneration is:

[0040] (9)

[0041] In the formula, is the minimum output power of the CHP unit; is the thermal power when power generation is minimum; is the maximum output power of CHP; is the heat generation power at time t; k min and k max It represents the electric-thermal conversion coefficient of the CHP unit corresponding to the minimum and maximum output power in the electric-thermal coupling constraint, Represents the linear slope of the electric power in the CHP unit; is the heat generation power of GB; is the natural gas consumption in GB, is the heat generation efficiency of GB; is the calorific value of natural gas, is the power generation efficiency of the CHP unit;

[0042] 2) Modeling of carbon capture device and power-to-gas device:

[0043] The CO2 emissions in the system come from the CHP unit and GB, so the CO2 emissions are modeled as:

[0044] (10)

[0045] Where, χ is the flue gas split ratio; ε is the carbon capture efficiency; , CO2 emissions from CHP units and GB;

[0046] The power-to-gas plant converts the captured CO2 into natural gas:

[0047] (11)

[0048] in, Expressed as the amount converted into natural gas; κ, are the electricity conversion efficiency during power-to-gas conversion and the proportional coefficient of the amount of CO2 required for P2G; ;

[0049] Therefore, to ensure stable operation of the system, each device should operate within its safety range:

[0050] (12)

[0051] 3) Other security constraints:

[0052] The amount of heat load removed satisfies the upper and lower limit constraints as follows:

[0053] (13)

[0054] The operation of the energy storage device meets the upper and lower limits of the charging and discharging power and the energy storage capacity:

[0055] (14)

[0056] In the formula, , It represents the charging and discharging power value of the energy storage in region i at time t; , Indicates the upper and lower limits of the energy storage charging power in area i; , Represents the upper and lower limits of the energy storage discharge power in region i; SoC i,t represents the remaining capacity of energy storage in region i at time t; represents the minimum allowable capacity of the energy storage device in region i at time t, represents the maximum permissible capacity of the energy storage equipment in region i at time t; , They represent the charging and discharging efficiency coefficients of energy storage respectively; Represents the time interval from time t-1 to time t; represents the capacity of energy storage in region i;

[0057] In actual operation, the output power of renewable energy operates within its maximum power range, and its constraints are as follows:

[0058] (15)

[0059] In the formula, , They represent the output power of photovoltaic and wind turbine in region i at time t, , They represent the maximum power of photovoltaic and wind turbine in region i at time t respectively.

[0060] Furthermore, step 4 is specifically as follows:

[0061] When a power shortage occurs in IES-i, if the power balance cannot be met by simply shedding the load, it is necessary to purchase energy from adjacent IES to achieve system balance;

[0062] To cooperate with other IES to promote system balance, the cost of establishing electric and heat exchange transmission between IES-i and IES-j is:

[0063] (16)

[0064] In the formula, , They represent the electrical and thermal powers interacting between IES-i and IES-j, respectively; , is the unit power price corresponding to the exchange of electricity and heat; accordingly, the supply and demand balance under the energy coordination between IES is changed to:

[0065] (17)

[0066] (18)

[0067] when , When , it means that the electric / thermal power flows from IES-i to IES-j. The power flows in the positive direction, indicating that the electric / thermal energy is sold. On the contrary, it flows from j to i, indicating that the electric / thermal energy is purchased. In actual operation, the electric / thermal power delivered from IES-i to IES-j is equal to the electric / thermal power received by IES-j from IES-i, that is:

[0068] (19)

[0069] The electrical and thermal energy transmission power output or received between IES at each time t should be within the transmission power limit of the tie line:

[0070] (20)

[0071] When a power shortage occurs in the distribution network, the stable operation of the system is promoted by cutting off a part of the interruptible power load and heat load and by the energy synergy between the IES. Therefore, the total operating cost of the system is:

[0072] (twenty one);

[0073] It can be seen from formula (21) that for an active distribution network consisting of multiple IES, the constraints of formula (19) lead to mutual coupling between the IES. In order to take into account the interests of different IES operating entities, it is necessary to perform a distributed solution to formula (21).

[0074] Furthermore, in step 5, for any IES-i in the main distribution network, the asynchronous ADMM algorithm is used to solve and realize the optimal scheduling strategy, which is specifically:

[0075] Step 5-1: The augmented Lagrange multiplier structure of the minimum operating cost objective function is as follows:

[0076] (twenty two),

[0077] In the formula, are the other corresponding Lagrange multipliers connected to IES-i, , ; , is the interaction energy between IES-i and IES-j; They represent the electrical and thermal powers interacting between IES-i and IES-j, respectively; is the Lagrangian parameter corresponding to IES-i, and its value is greater than 0;

[0078] Step 5-2: Initialize IES-i parameters , , , initialize the number of iterations ;

[0079] Step 5-3: For the kth IES-i i The iterations are as follows:

[0080] (twenty three);

[0081] Step 5-4, update the Lagrange multiplier to:

[0082] (twenty four);

[0083] Step 5-5: Update the amount in IES-i Send IES-j, j=1,2,…,N respectively, and send them to the dispatching center of the distribution network master station at the same time;

[0084] Step 5-6: When IES-i receives the updated value of the IES connected to it, it updates the mutual assistance variable, that is:

[0085] (25),

[0086] The updated value is sent to the distribution network master station dispatch center; otherwise, it is not updated;

[0087] Step 5-7: After receiving the update information of any other IES-j except IES-i, the distribution network master station updates the energy mutual aid variables between IES-i and IES-j, namely:

[0088] (26),

[0089] And send the updated value to the distribution network substation dispatch center;

[0090] Steps 5-8 , proceed to the next iteration;

[0091] Step 5-9: Check whether the convergence conditions are met, that is:

[0092] (27),

[0093] In the formula, Indicates the iteration accuracy termination parameter set in IES-i. When this condition is met, the iteration ends, otherwise continue with steps 5-3 to 5-9.

[0094] The beneficial effects described in the present invention are as follows: when a power shortage occurs in the system, the method described in the present invention adopts an asynchronous ADMM algorithm to solve the communication delay problem caused by the asynchronous data sampling in the thermal / electric coordination of each IES in a distributed manner. The computing nodes can be updated at different speeds and time points according to their own conditions without waiting for other nodes to synchronize, which greatly improves the flexibility and adaptability of the calculation, effectively handles the problem of information asynchrony, and improves the calculation efficiency of the algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] Figure 1 It is a schematic diagram of the multi-agent IES hierarchical distributed control strategy framework;

[0096] Figure 2 It is a schematic diagram of the asynchronous ADMM principle;

[0097] Figure 3 is the IES-1 power curve;

[0098] Figure 4 It is the IES-2 power curve;

[0099] Figure 5 It is the IES-3 power curve;

[0100] Figure 6 is the convergence curve of asynchronous ADMM;

[0101] Figure 7 It is a schematic diagram of the results of electric energy transactions between multiple IES when they are interconnected;

[0102] Figure 8 It is a schematic diagram of the comparison results between the asynchronous ADMM algorithm and the traditional ADMM algorithm in terms of convergence;

[0103] Fig. 9 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0104] In order to make the contents of the present invention more clearly understood, the present invention is further described in detail below based on specific embodiments in conjunction with the accompanying drawings.

[0105] like Fig. 9 As shown, the optimization scheduling strategy of a multi-agent integrated energy system under information asynchrony of the present invention comprises the following steps:

[0106] Step 1: Construct a multi-agent IES hierarchical distributed coordination control framework;

[0107] Step 2: Establish the IES operation objective function with the goal of minimizing the CHP unit operation cost, the cost of interacting with the external grid, the operation and maintenance cost of the energy storage system, and the load shedding cost;

[0108] Step 3: Based on the IES operation objective function, establish the IES supply and demand balance and system operation constraints;

[0109] Step 4: construct an electric / thermal cooperative mutual assistance model between multiple IES entities. When a power shortage occurs, the electric / thermal cooperative mutual assistance model between multiple IES entities is used to perform cooperative optimization, thereby achieving system balance and ensuring stable operation and efficiency optimization of the system.

[0110] Step 5: Based on the electric / thermal cooperative mutual assistance model among the multi-agent IES, an asynchronous ADMM algorithm is used to solve the problem, realize the optimal scheduling strategy, and solve the problem of information asynchrony.

[0111] In step 1, when there is a power shortage in the system operation, the system can be operated stably by complementing the flexible resources and new energy sources on the load side between different IES, giving full play to and mobilizing the energy complementarity characteristics between IES. In the IES studied in this invention, in addition to the combined heat and power (CHP) unit, it also includes a carbon capture system (CCS) and a power to gas (P2G) device. In actual operation, different IES belong to different operating entities, such as Figure 1 As shown in the figure, the IES Energy Management System (IES-EMS) can coordinate the internal heat / electric power of the IES. When the IES needs to coordinate heat / electricity with the IES connected to it, each IES-EMS needs to exchange some information with the upper-level distribution network dispatching center. The upper-level distribution network dispatching center will pass the updated variables to each IES-EMS to realize the distributed optimization dispatching of the IES.

[0112] In step 2, the IES operating cost is as follows:

[0113] (1)

[0114] In the formula, is the total cost when IES-i is operated alone. The right side of the equation includes the CHP unit operation cost, the cost of interacting with the external grid, the energy storage system operation and maintenance cost, and the load shedding cost.

[0115] (1) The operating cost of a CHP unit is:

[0116] (2)

[0117] Where a1, b1, c1 are the CHP unit operating cost coefficients. is the total output power of the CHP unit;

[0118] (2) Interaction costs with the external grid, including natural gas transaction costs and electricity transaction costs:

[0119] (3)

[0120] In the formula, , is the buy / sell gas price, , Indicates the amount of gas purchased and sold. , is the buying / selling price of electricity, , Indicates the electric power purchased from the power grid and the electric power sold to the large power grid;

[0121] (3) Considering that the service life of energy storage is limited, a certain amount of loss will occur during operation. The cost of the loss is expressed as:

[0122] (4);

[0123] (4) In order to maintain the safety of the system, when IES has a power shortage, it will interrupt part of the flexible load to ensure the balance between output and load. The cost paid to the interruptible load is:

[0124] (5)

[0125] In the formula, It represents the cost of users participating in shedding interruptible loads in IES-i; , are the compensation unit prices for cutting off interruptible load and thermal load in IES-i; δ i,t It represents the response coefficient of the interruptible load corresponding to the user participating in IES-i at time t, and this value is related to the user's willingness; , They represent the interruptible electrical load and thermal load power removed in IES-i respectively.

[0126] Step 3 is as follows:

[0127] (1) Electric power balance:

[0128] (6)

[0129] In the formula, , are the output power of photovoltaic and wind power respectively, The power supplied to the CHP unit for the electrical load, , Indicates the electric power consumed by CCS and P2G, Indicates the original electrical load demand;

[0130] (2) Thermal power balance:

[0131] (7)

[0132] In the formula, , Indicates the thermal power output of CHP units and gas boilers (GB). Indicates the original thermal power load;

[0133] (3) Gas balance constraints:

[0134] (8)

[0135] In the formula, is the natural gas consumption of the CHP unit, is the natural gas consumption in GB, The amount of natural gas produced for P2G;

[0136] (4) System safety operation constraints:

[0137] The normal operation of the system includes the combined heat and power unit CHP, the carbon capture device CCS, and the power-to-gas device P2G, and the modeling is as follows:

[0138] 1) The heat and power coupling output model of cogeneration is:

[0139] (9)

[0140] In the formula, is the minimum output power of the CHP unit; is the thermal power when power generation is minimum; is the maximum output power of CHP; is the heat generation power at time t; k min and k max It represents the electric-thermal conversion coefficient of the CHP unit corresponding to the minimum and maximum output power in the electric-thermal coupling constraint, Represents the linear slope of the electric power in the CHP unit; is the heat generation power of GB; is the natural gas consumption in GB, is the heat generation efficiency of GB; is the calorific value of natural gas, is the power generation efficiency of the CHP unit;

[0141] 2) Modeling of carbon capture device and power-to-gas device:

[0142] The CO2 emissions in the system come from the CHP unit and GB, so the CO2 emissions are modeled as:

[0143] (10)

[0144] Where, χ is the flue gas split ratio; ε is the carbon capture efficiency; , CO2 emissions from CHP units and GB;

[0145] The power-to-gas plant converts the captured CO2 into natural gas:

[0146] (11)

[0147] Among them, κ, are the electricity conversion efficiency during power-to-gas conversion and the proportional coefficient of the amount of CO2 required for P2G; ;

[0148] Therefore, to ensure stable operation of the system, each device should operate within its safety range:

[0149] (12);

[0150] 3) Other security constraints:

[0151] The amount of heat load removed satisfies the upper and lower limit constraints as follows:

[0152] (13)

[0153] The operation of the energy storage device meets the upper and lower limits of the charging and discharging power and the energy storage capacity:

[0154] (14)

[0155] In the formula, , It represents the charging and discharging power value of the energy storage in region i at time t; , represents the upper and lower limits of the energy storage charging power in area i; similarly, , Represents the upper and lower limits of the energy storage discharge power in region i; SoC i,t represents the remaining capacity of energy storage in region i at time t; , They represent the charging and discharging efficiency coefficients of energy storage respectively; Represents the time interval from time t-1 to time t; represents the capacity of energy storage in region i;

[0156] In actual operation, the output power of renewable energy operates within its maximum power range, and its constraints are as follows:

[0157] (15)

[0158] In the formula, , They represent the output power of photovoltaic and wind turbine in region i at time t, , They represent the maximum power of photovoltaic and wind turbine in region i at time t respectively.

[0159] Step 4 is as follows:

[0160] When a power shortage occurs in IES-i, if the power balance cannot be met by simply shedding the load, it is necessary to purchase energy from adjacent IES to achieve system balance;

[0161] To cooperate with other IES to promote system balance, the cost of establishing electric and heat exchange transmission between IES-i and IES-j is:

[0162] (16)

[0163] In the formula, , Respectively represent the electrical and thermal power exchanged between IES-i and IES-j; when the value is greater than 0, it means that the power flows from i to j, and vice versa; , is the unit power price corresponding to the exchange of electricity and heat; accordingly, the supply and demand balance under the energy coordination between IES is changed to:

[0164] (17)

[0165] (18)

[0166] when , When , it means that the electrical and thermal power flows from IES-i to IES-j, and vice versa. In actual operation, the electrical / thermal power delivered from IES-i to IES-j is equal to the electrical / thermal power received by IES-j from IES-i, that is:

[0167] (19)

[0168] The electrical and thermal energy transmission power output or received between IES at each time t should be within the transmission power limit of the tie line:

[0169] (20)

[0170] When a power shortage occurs in the distribution network, the stable operation of the system is promoted by cutting off a part of the interruptible power load and heat load and by the energy synergy between the IES. Therefore, the total operating cost of the system is:

[0171] (twenty one).

[0172] It can be seen from formula (21) that for an active distribution network consisting of multiple IES, the constraints of formula (19) lead to mutual coupling between the IES. In order to take into account the interests of different IES operating entities, it is necessary to perform a distributed solution to formula (21).

[0173] In step 5, different IES may have different calculation speeds and different data sampling frequencies when performing distributed optimization solutions. There is a communication delay problem between different IES in strategy calculation and update. If traditional ADMM is used to perform distributed calculation on formula (21), the calculation efficiency will be reduced. In order to solve the data information asynchrony caused by the above working conditions, the present invention adopts an asynchronous ADMM algorithm to realize the calculation of the scheduling strategy. Its principle is as follows: Figure 2 shown.

[0174] Taking IES-i as an example, the green bar indicates the calculation time of the optimization strategy in IES-i. The time schedule includes calculation speed, sampling period, etc. It can be seen that there are obvious differences between the various IES in the system. The distributed computing results of the IES-i dispatch center, including the independent variables in this area , draw a Langge multiplier and inter-regional energy mutual assistance variables When IES-i obtains the calculation result, it can immediately , It is sent to the connected IES and the dispatch center. Without loss of generality, it is assumed that i is connected to j and m. For an ADMM, as long as the dispatch center receives the update information of any area of ​​IES-j and IES-m, it can update the coupling variable information without receiving the information of all IES from the traditional ADMM algorithm. Assuming that the updated value of IES-m is received, the dispatch center can immediately update the energy mutual assistance variable between IES-i and IES-m, and send the updated value to the dispatch center. It is resent to the computing centers of i and m for the next iteration. It can be seen that in this method, as long as the update information of two IES arrives at the dispatch center, information interaction can be performed first, thereby reducing its waiting time. Here, when the dispatch center updates the energy mutual aid variable, the number of updates of IES cannot exceed two times, because the method proposed in the present invention mainly solves the distributed optimization scheduling strategy under limited delay caused by information asynchrony.

[0175] For any IES-i in the main distribution network, the asynchronous ADMM algorithm is used to solve and implement the optimal scheduling strategy, which is as follows:

[0176] (1) The augmented Lagrange multiplier structure of the minimum operating cost objective function is as follows:

[0177] (twenty two),

[0178] In the formula, are the other corresponding Lagrange multipliers connected to IES-i, , ; is the Lagrangian parameter corresponding to IES-i, and its value is greater than 0;

[0179] (2) Initialize IES-i parameters , , , initialize the number of iterations ;

[0180] (3) For the kth IES-i i The iterations are as follows:

[0181] (twenty three);

[0182] (4) Update the Lagrange multiplier to:

[0183] (twenty four);

[0184] (5) Update the quantity in IES-i Send IES-j, j=1,2,…,N respectively, and send them to the dispatching center of the distribution network master station at the same time;

[0185] (6) When IES-i receives the updated value of the IES connected to it, it updates the mutual assistance variable, that is:

[0186] (25),

[0187] The updated value is sent to the distribution network master station dispatch center; otherwise, it is not updated;

[0188] (7) After the distribution network master station receives the update information of any other IES-j other than IES-i, it updates the energy mutual benefit variables between IES-i and IES-j, that is:

[0189] (26),

[0190] And send the updated value to the distribution network substation dispatch center;

[0191] (8) ;

[0192] (9) Check whether the convergence conditions are met, that is:

[0193] (27),

[0194] In the formula, It represents the iteration accuracy termination parameter set in IES-i. When this condition is met, the iteration ends; otherwise, continue with steps (3) to (9).

[0195] In order to verify the effectiveness of the multi-agent IES load shedding strategy proposed in the present invention, the three IES are verified through simulation data. At the same time, each IES is connected to each other and to the large power grid. Its structure is as follows: Figure 1 shown.

[0196] In IES-1 and IES-2, the renewable power generation units include wind turbines and conventional CHP units. In IES-3, a combined heat and power generator with integrated carbon capture and power-to-gas is included, and the renewable power generation unit is photovoltaic. The power operating range of the CHP unit in the calculation example is [1200, 5000] kW, η CHP =0.35, the electrical-thermal coupling conversion coefficients are k min =0.15, k max = 0.2, k m =0.85; the parameters of the CHP unit including carbon capture and power-to-gas are: The power range is [400, 2400] kW, The power range is [0, 600] kW, 2000 kW·h, the SoC operation range is [0.2, 0.9], the maximum power of charge and discharge is 500 kW and 600 kW respectively, and the limit value of energy interaction between IES is 5000 kW; the gas price is 0.3 yuan / m 3 The electricity price between IES is 0.35 yuan / kW·h. Figure 3-Figure 5 The output power and electrical / thermal load curves of renewable energy in the three IES are given respectively.

[0197] Figure 6 The optimization scheduling results are given when IES-2 is only connected to the distribution network. It can be seen that in order to ensure the stable operation of the system, the CHP unit of IES-2 is always in operation and occupies most of the energy supply, followed by purchasing electricity from the large power grid, which will increase the system operation cost.

[0198] When multiple IES are interconnected, the results of the power transactions between them are as follows: Figure 7 As shown, Figure 7 A positive value indicates the energy selling state, and a negative value indicates the energy purchasing state. It can be seen that the IES achieves stable system operation through energy complementation. For example, Figure 5 It can be seen from the figure that at the initial time 1-6, the photovoltaic output power of IES-3 is 0. Therefore, in addition to supplying power to meet the load demand, IES-3 purchases energy from IES-1 and IES-2 to achieve supply and demand balance. Since the electricity price between IESs is lower than the power grid price, the operating cost of IES-3 is reduced through interconnection.

[0199] According to the simulation case calculation results of the present invention, the operating costs of IES-1, IES-2, and IES-3 decreased by 3.05%, 4.7%, and 19.50%, respectively. The specific values ​​are shown in Table 1. The cost reduction of IES-3 is most obvious because the power shortage in IES-3 is large, and the price of purchasing electricity from IES-1 and IES-2 is significantly lower than the real-time grid electricity price. Therefore, the operating cost of IES-3 is significantly reduced in the interconnected operation mode.

[0200] Table 1 Comparison of operating costs under the collaborative mode between IES

[0201]

[0202] The asynchronous ADMM algorithm is used to solve the electrical / thermal dual-layer distributed optimization scheme between different entities, and is compared with the traditional ADMM algorithm. The parameters of the two algorithms are set the same in the simulation: the maximum number of iterations is 100, the penalty parameter is set to 0.1, and the residual value is set to 10 -3 Its computational efficiency performance is shown in Table 2. From the simulation results, it can be seen that the asynchronous ADMM requires less time in the iteration process. This is because the asynchronous ADMM needs less waiting time when updating the Lagrangian factor. The asynchronous ADMM algorithm is nearly 40% more efficient than the ADMM algorithm, so its efficiency is significantly better than the traditional ADMM algorithm.

[0203] Table 2 Comparison of computational efficiency of different algorithms

[0204]

[0205] At the same time, the present invention compares the convergence of the asynchronous ADMM algorithm with the traditional ADMM algorithm, and the results are as follows: Figure 8 As shown, when the residual value is less than the set 10 -3 When , the program terminates and the algorithm converges. From the simulation results, it can be seen that both methods can achieve good convergence effects. The asynchronous ADMM requires more iterations than the traditional ADMM algorithm. This is because when only two IES update information is transmitted to the distribution network master station dispatching center, the Lagrangian parameters will be updated, so more iterations are required to complete the convergence of the algorithm.

[0206] In the actual simulation setting, the information iteration between different IES only requires the cooperative energy parameters between the interactive adjacent regions without other data. When the latest updated value in the adjacent region has not arrived, IES-EMS will update the Lagrange multiplier according to the last iteration value, as shown in formula (26), which improves the efficiency of each IES in the energy strategy calculation.

[0207] Aiming at the problem of information asynchrony in the cooperation between multiple IES entities when the system has power shortage, the present invention proposes a solution based on asynchronous ADMM distributed algorithm, and realizes the stable operation of the system through the thermal / electric power cooperation between IES entities. The method proposed in the present invention is verified through specific cases. Simulation results show that the method proposed in the present invention can effectively improve the computing efficiency when the information is asynchrony, and the cooperation between IES entities can reduce the system operation cost.

[0208] The above description is only a preferred embodiment of the present invention and is not intended to be a further limitation of the present invention. All equivalent changes made using the contents of the present specification and drawings are within the protection scope of the present invention.

Claims

1. A method for optimizing and dispatching a multi-agent integrated energy system under information asynchrony, characterized in that: The following steps are involved: Step 1: Construct a multi-agent IES hierarchical distributed coordination control framework; Step 2: Establish the IES operation objective function with the goal of minimizing the CHP unit operation cost, the cost of interacting with the external grid, the operation and maintenance cost of the energy storage system, and the load shedding cost; Step 3: Based on the IES operation objective function, establish the IES supply and demand balance and system operation constraints; Step 4: construct an electric / thermal cooperative mutual assistance model between multiple IES entities. When a power shortage occurs, the electric / thermal cooperative mutual assistance model between multiple IES entities is used to perform cooperative optimization to achieve system balance. Specifically: When a power shortage occurs in IES-i, if the power balance cannot be met by simply shedding the load, it is necessary to purchase energy from adjacent IES to achieve system balance; To cooperate with other IES to promote system balance, the cost of establishing electric and heat exchange transmission between IES-i and IES-j is: (16)， Where, N IES, T is the total operation time of the CHP unit, and t is a certain moment of the CHP unit operation; , They represent the electrical and thermal powers interacting between IES-i and IES-j, respectively; , is the unit power price corresponding to the exchange of electricity and heat; accordingly, the supply and demand balance under the energy coordination between IES is changed to: (17)， (18)， in, , Indicates the electric power purchased from the power grid and the electric power sold to the large power grid; , Respectively represent the charging and discharging power of the energy storage system; , are the output power of photovoltaic and wind power respectively, The power supplied to the CHP unit for the electrical load, , Indicates the electric power consumed by CCS and P2G, Indicates the original electrical load demand; , They represent the interruptible electrical load and thermal load power removed in IES-i respectively; , Indicates the thermal power output of the CHP unit and gas boiler GB. Indicates the original thermal power load; when , When , it means that the electrical and thermal power flows from IES-i to IES-j, and vice versa. In actual operation, the electrical and thermal power delivered from IES-i to IES-j is equal to the electrical and thermal power received by IES-j from IES-i, that is: (19)， The electrical and thermal energy transmission power output or received between IES at each time t should be within the transmission power limit of the tie line: (20)， When a power shortage occurs in the distribution network, the stable operation of the system is promoted by cutting off a part of the interruptible power load and heat load and by the energy synergy between the IES. Therefore, the total operating cost of the system is: (21)； in, is the total cost when IES-i is operated alone; Step 5: Based on the electric / thermal cooperative mutual assistance model among the multi-agent IES, an asynchronous ADMM algorithm is used to solve and realize the optimal scheduling strategy.

2. According to the method for optimizing and dispatching a multi-agent integrated energy system under information asynchrony condition described in claim 1, it is characterized in that: Step 2 is as follows: Assuming there are N IES, the operating cost of IES-i is as shown in formula (1), where 1≤i≤N: (1)， In the formula, is the total cost of IES-i when running alone, is the CHP unit operating cost, The cost of interacting with the external network is is the operation and maintenance cost of the energy storage system, Costs for users in IES-i to participate in shedding interruptible loads; (1) The operating cost of a CHP unit is: (2)， Where a1, b1, c1 are the CHP unit operating cost coefficients. is the total output power of the CHP unit, t is a certain moment when the CHP unit is running, and T is the total operating time of the CHP unit; (2) Interaction costs with the external grid, including natural gas transaction costs and electricity transaction costs: (3)， In the formula, , is the buy / sell gas price, , Indicates the amount of gas purchased and sold. , is the buying / selling price of electricity, , Indicates the electric power purchased from the power grid and the electric power sold to the large power grid; (3) Considering that the service life of energy storage is limited, a certain amount of loss will occur during operation. The cost of the loss is expressed as: (4)， In the formula , They represent the charging and discharging maintenance coefficients of the energy storage system in IES-i respectively; , Respectively represent the charging and discharging power of the energy storage system; (4) In order to maintain the safety of the system, when IES has a power shortage, it will interrupt part of the flexible load to ensure the balance between output and load. The cost paid to the interruptible load is: (5)， In the formula, , They are the compensation unit prices for cutting off interruptible load and thermal load in IES-i respectively; , They represent the interruptible electrical load and thermal load power removed in IES-i respectively.

3. According to the method for optimizing and dispatching a multi-agent integrated energy system under information asynchrony condition described in claim 2, the characteristics are: Step 3 is as follows: (1) Electric power balance: (6)， In the formula, , are the output power of photovoltaic and wind power respectively, The power supplied to the CHP unit for the electrical load, , Indicates the electric power consumed by CCS and P2G, Indicates the original electrical load demand; (2) Thermal power balance: (7)， In the formula, , Indicates the thermal power output of the CHP unit and gas boiler GB. Indicates the original thermal power load; (3) Gas balance constraints: (8)， In the formula, is the natural gas consumption of the CHP unit, is the natural gas consumption in GB, The amount of natural gas produced for P2G; (4) System safety operation constraints: The normal operation of the system includes the combined heat and power unit CHP, the carbon capture device CCS, and the power-to-gas device P2G, and the modeling is as follows: 1) The heat and power coupling output model of cogeneration is: (9)， In the formula, is the minimum output power of the CHP unit; is the thermal power when power generation is minimum; is the maximum output power of CHP; is the heat generation power at time t; k min and k max It represents the electric-thermal conversion coefficient of the CHP unit corresponding to the minimum and maximum output power in the electric-thermal coupling constraint, Represents the linear slope of the electric power in the CHP unit; is the heat generation power of GB; is the natural gas consumption in GB, is the heat generation efficiency of GB; is the calorific value of natural gas, is the power generation efficiency of the CHP unit; 2) Modeling of carbon capture device and power-to-gas device: The CO2 emissions in the system come from the CHP unit and GB, so the CO2 emissions are modeled as: (10)， Where, χ is the flue gas split ratio; ε is the carbon capture efficiency; , CO2 emissions from CHP units and GB; The power-to-gas plant converts the captured CO2 into natural gas: (11)， in, Expressed as the amount converted into natural gas; κ, are the electricity conversion efficiency during power-to-gas conversion and the proportional coefficient of the amount of CO2 required for P2G; ; Therefore, to ensure stable operation of the system, each device should operate within its safety range: (12)； 3) Other security constraints: The amount of heat load removed satisfies the upper and lower limit constraints as follows: (13)， The operation of the energy storage device meets the upper and lower limits of the charging and discharging power and the energy storage capacity: (14)， In the formula, , It represents the charging and discharging power value of the energy storage in region i at time t; , Indicates the upper and lower limits of the energy storage charging power in area i; , Represents the upper and lower limits of the energy storage discharge power in region i; SoC i,t represents the remaining capacity of energy storage in region i at time t; represents the minimum allowable capacity of the energy storage device in region i at time t, represents the maximum permissible capacity of the energy storage equipment in region i at time t; , They represent the charging and discharging efficiency coefficients of energy storage respectively; Represents the time interval from time t-1 to time t; represents the capacity of energy storage in region i; In actual operation, the output power of renewable energy operates within its maximum power range, and its constraints are as follows: (15)， In the formula, , They represent the output power of photovoltaic and wind turbine in region i at time t, , They represent the maximum power of photovoltaic and wind turbine in region i at time t respectively.

4. According to claim 1, a method for optimizing and dispatching a multi-agent integrated energy system under information asynchrony is characterized in that: In step 5, for any IES-i in the main distribution network, the asynchronous ADMM algorithm is used to solve and implement the optimal scheduling strategy, which is: Step 5-1: The augmented Lagrange multiplier structure of the minimum operating cost objective function is as follows: (22)， In the formula, are the other corresponding Lagrange multipliers connected to IES-i, , ; , is the interaction energy between IES-i and IES-j; They represent the electrical and thermal powers interacting between IES-i and IES-j, respectively; is the Lagrangian parameter corresponding to IES-i, and its value is greater than 0; Step 5-2: Initialize IES-i parameters , , , initialize the number of iterations ; Step 5-3: For the kth IES-i i The iterations are as follows: (23)； Step 5-4, update the Lagrange multiplier to: (24)； Step 5-5: Update the amount in IES-i Send IES-j, j=1,2,…,N respectively, and send them to the dispatching center of the distribution network master station at the same time; Step 5-6: When IES-i receives the updated value of the IES connected to it, it updates the mutual assistance variable, that is: (25)， The updated value is sent to the distribution network master station dispatch center; otherwise, it is not updated; Step 5-7: After receiving the update information of any other IES-j except IES-i, the distribution network master station updates the energy mutual aid variables between IES-i and IES-j, namely: (26)， And send the updated value to the distribution network substation dispatch center; Steps 5-8 , proceed to the next iteration; Step 5-9: Check whether the convergence conditions are met, that is: (27)， In the formula, Indicates the iteration accuracy termination parameter set in IES-i. When this condition is met, the iteration ends, otherwise continue with steps 5-3 to 5-9.

Images