Large-scale energy storage power station optimization power distribution method considering dynamic reconfigurable battery output constraint

Abstract

The invention discloses a large energy storage power station optimization power distribution method considering dynamic reconfigurable battery output constraint, which is suitable for power scheduling in a dynamic reconfigurable battery energy storage power station and solves the problems of single target, poor cost effectiveness and the like of the existing power distribution method. According to the method, multiple target factors such as the operation and maintenance cost of the battery energy storage unit, the battery loss cost and the consistency of the state of charge of the battery of each energy storage unit are combined, the power constraint of the dynamic reconfigurable battery on the inverter unit is considered, and the weight problem of different targets is solved by adopting a fuzzy set theory and a multivariate comparison weight determination method. According to the method, the consistency of the charge states of the energy storage batteries can be effectively enhanced, the overall operation efficiency and reliability of the battery energy storage power station are improved, and meanwhile the cost effectiveness of the energy storage power station is improved through the optimized power distribution method.

Description

Technical Field

[0001] The present invention relates to the technical field of power dispatching of battery energy storage power stations, and in particular to a method for optimizing power allocation of large energy storage power stations taking into account dynamic reconfigurable battery output constraints. Background Art

[0002] In the process of energy transformation, battery energy storage technology is widely used to solve problems such as discontinuity, unpredictability and instability of renewable energy output. Among them, dynamic reconfigurable battery technology, as a technology that can adjust battery output constraints in real time, allows battery energy storage systems to flexibly reconstruct the output power and battery operating mode of the battery pack according to load demand and environmental changes. This technology can not only improve the efficiency of battery charging and discharging, but also avoid over-discharge or over-charging by adjusting the battery's operating mode, reduce battery life loss, and optimize the battery's service life cycle. At the same time, the dynamic reconfigurable battery can also dynamically adjust the power output according to the real-time health status of the battery pack, thereby ensuring the stability and long-term economic benefits of the energy storage power station.

[0003] The rapid development of dynamically reconfigurable battery energy storage power stations has led to their widespread use in energy supply on the power supply side or the load side. In previous studies on power station power allocation strategies, most energy storage power stations mainly focused on operating costs as a single goal. This strategy can ensure the short-term benefits of energy storage power stations, but it is easy to cause the consistency of the battery energy storage state of charge to decrease, shorten the service life of energy storage batteries and units, and affect the long-term benefits of energy storage power stations. In addition, the single-objective energy allocation strategy has a low reference value in practical applications, while the multi-objective allocation strategy can enhance the overall consistency within the battery energy storage power station while meeting the economic needs of the power station, ensuring long-term sustainable development. Therefore, it is particularly important to design an optimized power allocation method for large energy storage power stations that considers the dynamic reconfigurable battery output constraints. Summary of the invention

[0004] In order to solve the problems that the power allocation strategy of battery energy storage power stations is mostly based on cost as a single goal, resulting in high loss cost and poor consistency of battery energy storage units in the power station, the purpose of the present invention is to design a large-scale energy storage power station optimization power allocation method considering dynamic reconfigurable battery output constraints.

[0005] In order to achieve the above object, the present invention adopts the following technical solution:

[0006] A method for optimizing power allocation of a large energy storage power station considering dynamic reconfigurable battery output constraints comprises the following steps:

[0007] S1: Establish an objective function model that takes into account the operation and maintenance costs of battery energy storage units to improve the operational reliability of large energy storage power stations;

[0008] S2: Establish an objective function model that takes into account the battery loss cost of the battery energy storage unit;

[0009] S3: Establish an objective function model that takes into account the temperature regulation cost of the battery energy storage unit to ensure the safety and economy of the battery energy storage unit;

[0010] S4: Establish an objective function model that considers the consistency of battery state of charge parameters of the battery energy storage unit to enhance the overall performance of the battery energy storage unit;

[0011] S5: Based on the objective function model established in the above steps S1-S4, the weight of each objective function is designed using fuzzy set theory and multivariate comparison weight determination method;

[0012] S6: Consider the output constraints of the dynamically reconfigurable battery, combine the maximum output power of the inverter with the voltage of the DC side battery pack, and design the constraints;

[0013] S7: Based on the known state parameters of the battery energy storage unit, an intelligent algorithm is used to solve the multiple objective function models established in steps S1-S4 to achieve power distribution of the energy storage power station.

[0014] The specific method of step S1 is as follows:

[0015] S1.1: Assume that the operation and maintenance cost coefficient of the i-th battery energy storage unit in the battery energy storage power station is The operation and maintenance loss of a single battery energy storage unit is:

[0016]

[0017] Where P i is the output power of the i-th battery energy storage unit.

[0018] S1.2: Assuming that the output power is constant and n battery energy storage units work for T hours, the objective function model of the operation and maintenance cost of the battery energy storage unit is established as follows:

[0019]

[0020] The specific method of step S2 is as follows:

[0021] Assume that the power loss coefficient of the i-th battery energy storage unit in the battery energy storage power station is The discharge efficiency of the battery of the i-th energy storage unit is η 1 Then the objective function model of battery loss cost of battery energy storage unit can be established as:

[0022]

[0023] Where P iis the output power of the i-th energy storage unit, T is the working time, and n is the number of battery energy storage units.

[0024] The specific method of step S3 is as follows:

[0025] S3.1: Consider that the heat generation power is related to the output power, where the heat generation power includes the heat generation power of the battery and the heat generation power of the inverter;

[0026] Q A (i) = Q B (i)+Q C (i) (4)

[0027] In the formula, Q A (i) is the total heat generation power of the battery energy storage unit, Q B (i) is the heat generation power of the battery, Q C (i) is the heat generation power of the inverter;

[0028]

[0029] Where P dis is the discharge power of the battery energy storage unit, η 1 is the discharge efficiency of the energy storage battery, η 2 is the output efficiency of the inverter;

[0030] S3.2: Assuming the output power is constant and considering n battery energy storage units working for t hours, the objective function model of temperature regulation cost is established as:

[0031]

[0032] C A (i) = P A (i)×P pri ×t

[0033] In the formula, K is the ratio of heat generation or heat removal power to electrical power, P pri is the electricity purchase price, C A (i) The cost required for temperature regulation.

[0034] The specific method of step S4 is as follows:

[0035] S4.1: Considering the state of charge of the energy storage battery after power output, the update formula is:

[0036] SOC new =SOC i -((P i *T) / E max ) (7)

[0037] In the formula, SOCi is the state of charge of the battery of the i-th battery energy storage unit, E max is the maximum capacity of the energy storage battery, T is the working time of the battery energy storage unit, SOC new It is the state of charge of the energy storage battery after power output.

[0038] S4.2: Establish an objective function model that considers the consistency of battery state of charge parameters of the battery energy storage unit:

[0039]

[0040] In the formula, SOC avg It is the average value of the state of charge of the energy storage battery after power output.

[0041] The specific method of step S5 is as follows:

[0042] S5.1: Referring to the fuzzy set theory, the membership function F is used to describe the optimization results of each objective function. The membership function F is defined as follows:

[0043]

[0044] In the formula, F i (X) is the target f i The membership function of (X), X i * is the optimal strategy for target optimization, f iw It is the worst value of the objective when optimizing each single objective.

[0045] S5.2: Convert each objective function model established in steps S1-S4 into a corresponding membership function, and obtain the final objective function model:

[0046]

[0047] S5.3: Use the multivariate comparison weighting method to determine the fuzzy scale of each target, convert the fuzzy mood operator into the corresponding membership degree, and normalize it to obtain the weight coefficient w i .

[0048] Compared with the existing method, the beneficial effects of the present invention are:

[0049] The present invention combines factors such as the state of charge of the energy storage battery and the output power of the energy storage inverter PCS to design a multi-objective active power allocation method that takes into account the dynamic reconfigurable battery output constraints to optimize the power allocation of energy storage units inside large battery energy storage power stations. The output constraints of the dynamically reconfigurable battery are added to the power allocation method, and the particle swarm algorithm based on fuzzy set theory is used to solve the model to achieve efficient power allocation. This method not only ensures the economic benefits of the battery energy storage power station, but also ensures the consistency of the state parameters of the battery energy storage units inside the power station, thereby improving the overall efficiency. In addition, in the process of solving the multi-objective function model, the present invention adopts a multivariate comparative weighting method to reduce the impact caused by the differences in dimensions and weights between multiple objectives. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is a typical 3S architecture diagram of a battery energy storage power station;

[0051] Figure 2 It is the circuit topology diagram of the bidirectional DC / AC converter;

[0052] Figure 3 The power allocation results comparison chart of the optimization strategy and the equal distribution strategy;

[0053] Figure 4 It is a comparison chart of the state of charge of the optimization strategy and the equalization strategy before and after power output;

[0054] Figure 5 This is the flow chart of the particle swarm algorithm based on fuzzy set theory. DETAILED DESCRIPTION

[0055] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0056] The present invention proposes a method for optimizing power allocation of a large energy storage power station considering dynamic reconfigurable battery output constraints. The method considers multiple target factors such as the state of charge of the battery energy storage unit, the operation and maintenance cost of the energy storage unit, and the battery loss cost, and adds the output constraint of the dynamic reconfigurable battery to the power allocation method. On the basis of ensuring economic costs, the method improves the overall consistency of the state of charge between different units in the battery energy storage power station and ensures the efficient operation of the battery energy storage power station.

[0057] Large energy storage power stations usually adopt a 3S architecture consisting of a battery management system, a power conversion system, and an energy management system. The three work together to convert the battery energy into power according to the grid demand, and manage the battery charging and discharging process through corresponding protection control logic to ensure the safe operation of the power station. The typical structure of a large energy storage power station is shown in the figure below. Figure 1 shown.

[0058] The main component of the power conversion system is the converter, which can adjust the power characteristics such as voltage, frequency and number of phases. In the converter (PCS), the commonly used converter types include bidirectional DC / DC converter and bidirectional DC / AC converter. Bidirectional DC / DC converter is mainly used to adjust the DC voltage, while bidirectional DC / AC converter realizes the energy interaction between the energy storage power station and the power grid. Figure 2 The circuit topology of the bidirectional DC / AC converter is shown.

[0059] according to Figure 1 The structural diagram of a large energy storage power station can be assumed that the total output power command given by the energy storage power station is P, and there are n energy storage units in the power station. In the process of model design and solution, the battery state parameters and total power command in the energy storage power station are known. The steps of a large energy storage power station optimization power allocation method considering dynamic reconfigurable battery output constraints are as follows:

[0060] 1) Establish an objective function model that considers the operation and maintenance costs of battery energy storage units:

[0061] Assume that the operation and maintenance cost coefficient of the i-th battery energy storage unit in the battery energy storage power station is The operation and maintenance loss of a single battery energy storage unit is:

[0062]

[0063] Where P i is the output power of the i-th battery energy storage unit.

[0064] Operation and maintenance cost coefficient It reflects the maintenance and operating costs per unit power of the battery energy storage unit during operation. This coefficient not only takes into account factors such as battery performance degradation, maintenance frequency, and operating conditions, but also integrates the impact of the external environment (such as climate and load fluctuations) on equipment maintenance costs. Different units may have different operation and maintenance cost coefficients due to differences in technology, scale, load characteristics, etc. Therefore, it is necessary to set a reasonable It is the key to optimizing the economic efficiency of energy storage systems.

[0065] Assuming that the output power is constant and considering that n battery energy storage units work for T hours, the objective function of the operation and maintenance cost of the battery energy storage unit is established as:

[0066]

[0067] 2) Establish an objective function model that takes into account the design battery loss cost:

[0068] Assume that the power loss coefficient of the i-th battery energy storage unit in the battery energy storage power station is The discharge efficiency of the battery in the i-th battery energy storage unit is η1 Then the objective function of the battery loss cost in the battery energy storage unit can be established as:

[0069]

[0070] Where P i is the output power of the i-th battery energy storage unit, T is the working time of the battery energy storage unit, and n is the number of battery energy storage units.

[0071] Battery loss factor It indicates the energy loss caused by internal resistance, battery chemical reaction, temperature and other factors during the operation of the battery energy storage unit under the condition of unit power output. This coefficient is affected by many factors, such as battery technology type, charge and discharge cycle, service life and ambient temperature. It can accurately reflect the loss characteristics of the battery energy storage unit under different loads, which helps to optimize the energy efficiency and economy of the battery energy storage unit.

[0072] 3) Establish an objective function model considering the temperature regulation cost of the battery energy storage unit:

[0073] Consider that the heat generation power is related to the output power, including the heat generation power of the battery and the heat generation power of the inverter.

[0074] Q A (i) = Q B (i)+Q C (i) (4)

[0075] In the formula, Q A (i) is the total heat generation power of the battery energy storage unit, Q B (i) is the heat generation power of the battery, Q C (i) is the heat generation power of the inverter.

[0076]

[0077] Where P dis is the discharge power of the battery energy storage unit, η 1 is the discharge efficiency of the energy storage battery, η 2 is the output efficiency of the inverter.

[0078] Assuming the output power is constant and considering that n battery energy storage units work for t hours, the objective function model of temperature regulation cost is:

[0079]

[0080] C A (i) = P A (i)×P pri ×t

[0081] In the formula, K is the ratio of heat generation or heat removal power to electrical power, P pri is the electricity purchase price, C A (i) The cost required for temperature regulation.

[0082] 4) Establish an objective function model that considers the consistency of battery charge state in the battery energy storage unit:

[0083] To ensure that the state of charge of the energy storage battery is consistent after power output, the state of charge update after power output is considered during design. The formula is:

[0084] SOC new =SOC i -((P i *T) / E max ) (7)

[0085] In the formula, SOC i is the state of charge of the battery of the i-th battery energy storage unit, E max is the maximum capacity of the energy storage battery, T is the working time of the battery energy storage unit, SOC new It is the state of charge of the energy storage battery after power output.

[0086] Establish an objective function model that considers the consistency of battery state of charge parameters of battery energy storage units:

[0087]

[0088] In the formula, SOC avg It is the average value of the state of charge of the energy storage battery after power output.

[0089] 5) Use fuzzy set theory to linearly transform the above multiple objectives:

[0090] Since the dimensions and weights of the above four objective function models are different, according to the fuzzy set theory, the membership function F is used to describe the optimization results of each objective function model, and the membership value range is 0 to 1. Specifically, F = 0 means that the target optimization degree is the highest, the closest to the optimal result, and the optimization result is the most satisfactory; while F = 1 means that it is farthest from the optimal result and the optimization result is the worst.

[0091] The membership function F is defined as follows:

[0092]

[0093] In the formula, F i (X) is the target f i(X) membership function, for the convenience of transformation, all objective functions are uniformly expressed as minimization problems (for the objective function of maximizing economic benefits, it is converted into minimization form by taking the negative sign); X i * For the target f i (X) The optimal strategy during optimization, f iw is the worst value of the target when optimizing each single target, that is,

[0094] By using the linear weighted summation method, the designed multi-objective optimization problem is transformed into a single-objective problem, and the following objective function F is obtained:

[0095]

[0096] In the formula, w i is the weight coefficient, satisfying the following relationship:

[0097]

[0098] 6) Dynamically reconfigurable battery output constraints:

[0099] Since the output power of the inverter unit will be affected by the DC side voltage, the range constraint of the dynamic reconfigurable battery pack state on the inverter output power is considered during the design, that is, the maximum output power of the inverter is constrained by the dynamic reconfigurable battery pack state.

[0100] The state of the dynamically reconfigurable battery determines the voltage on the DC side. Different reconfiguration states correspond to different voltage values. Taking the DC side voltage of the inverter as Vdc and the maximum output current of the inverter as fixed, the maximum output power of the inverter can be obtained as:

[0101] Pmax=Vdc*Imax / 1000 (12)

[0102] Where Imax is the maximum output current of the inverter.

[0103] 7) Design of weight coefficient:

[0104] The weight coefficient is determined by using the comparative weighting method of qualitative sorting and quantitative scaling. The basic steps are as follows:

[0105] For an indicator set consisting of n indicators {F 1 ,F 2 ,…,F n}, firstly, the importance of each indicator is compared in binary. If the indicator F k Than F l More important, then set the sorting scale e kl =1,e lk=0; otherwise, if F l More important, set e kl =0,e lk =1; if F k and F l Equally important, set e kl =e lk =0.5. Based on these comparison results, the importance qualitative ranking scale matrix E is constructed:

[0106] E=(e kl ) n×n (13)

[0107] Matrix E needs to be checked for consistency to ensure there are no logical contradictions. The specific rules are as follows:

[0108] If e hk >e hl , then there should be e lk >e kl ;

[0109] If e hk <e hl , then there should be e lk <e kl ;

[0110] If e hk =e hl =0.5, then e lk =e kl =0.5.

[0111] Then, the sum of the elements in each row of the matrix E is calculated, and the qualitative ranking of the indicator set is determined according to the order of their size. By comparing different indicators, the importance of each indicator is described using fuzzy language and text, and then these fuzzy tones are converted into corresponding membership values, normalized, and the weight coefficient w of the indicator is obtained. i .

[0112] Assuming that the power output of the energy storage power station is P, and n battery energy storage units are dispatched to complete the dispatching task, the particle swarm algorithm based on fuzzy set theory is used to solve the power requirement of each unit, such as Figure 3 At the same time, the solution results of the method of the present invention are compared with the solution results of the power sharing method to obtain a comparison diagram of the battery state of charge parameters after power output, as shown in Figure 4 As shown, it can be seen from the figure that compared with the power sharing method, after the output power is solved by the method of the present invention, the charge state consistency of the energy storage battery is improved and the integrity of the battery energy storage unit is enhanced.

[0113] In addition, by comparing the costs of the power optimization allocation method proposed in the present invention with those of the equal power sharing method, it can be found that the power optimization allocation method proposed in the present invention has good benefits in terms of the cost of power station power scheduling, ensuring the economy of the battery energy storage power station.

[0114] In this embodiment, as Figure 5 shown, the particle swarm algorithm based on the fuzzy set theory is used to solve the multi-objective function model, and the specific steps are as follows:

[0115] ① Solve the optimal solution and the worst solution of each designed single-objective function model;

[0116] ② Initialize the particle swarm, including the population size N, the position x i of each particle and the velocity v i ;

[0117] ③ Determine the weight coefficient by using the multi-element comparison and weight determination method;

[0118] ④ Calculate the membership degree of each particle and normalize it;

[0119] ⑤ Calculate the fuzzy objective function value Fit(i) of each particle, and its formula is as follows:

[0120]

[0121] ⑥ For each particle, compare its fuzzy objective function value Fit(i) with the individual extreme value pbest(i). If Fit(i) < pbest(i), then update the value of pbest(i);

[0122] ⑦ For each particle, compare its fuzzy objective function value Fit(i) with the global extreme value gbest. If Fit(i) < gbest, then update the value of gbest;

[0123] ⑧ Iteratively update the velocity v i and the position x i ;

[0124] ⑨ Perform boundary condition processing, that is:

[0125] v min ≤v i ≤v max (15)

[0126] xmin ≤ xi ≤ xmax

[0127] ⑩ Judge whether the algorithm termination condition is satisfied. If it is satisfied, stop the operation and output the optimization result; otherwise, return to step ⑤ and repeat this process until it ends.

Claims

1. A method for optimizing power allocation of a large energy storage power station considering dynamic reconfigurable battery output constraints, characterized in that: It includes the following steps: S1: Establish an objective function model considering the operation and maintenance costs of the battery energy storage unit to improve the operation reliability of the large-scale energy storage power station; S2: Establish an objective function model considering the battery loss cost of the battery energy storage unit; S3: Establish an objective function model considering the temperature regulation cost of the battery energy storage unit to ensure the safety and economy of the battery energy storage unit; S4: Establish an objective function model considering the consistency of the state of charge parameters of the battery energy storage unit to enhance the overall performance of the battery energy storage unit; S5: Integrate the objective function models established in the above steps S1 - S4, and use the fuzzy set theory and the multiple comparison and weight determination method to design the weights of each objective function; S6: Considering the output constraints of the dynamically reconfigurable battery, combine the maximum output power of the inverter with the voltage of the battery pack on the DC side to design the constraint conditions; S7: According to the known state parameters of the battery energy storage unit, use an intelligent algorithm to solve the multiple objective function models established in steps S1 - S4 to achieve the power distribution of the energy storage power station.

2. According to claim 1, a method for optimizing power allocation of a large energy storage power station considering dynamic reconfigurable battery output constraints is characterized in that: The specific method of step S1 is as follows: S1.1: Assume that the operation and maintenance cost coefficient of the i-th battery energy storage unit in the battery energy storage power station is The operation and maintenance loss of a single battery energy storage unit is: Where P i is the output power of the i-th battery energy storage unit; S1.2: Assume that the output power is constant. Considering that n battery energy storage units work for T hours, the objective function model of the operation and maintenance cost of the battery energy storage unit is established as:

3. According to claim 1, a method for optimizing power allocation of a large energy storage power station considering dynamic reconfigurable battery output constraints is characterized in that: The specific method of step S2 is as follows: Assume that the power loss coefficient of the i-th battery energy storage unit in the battery energy storage power station is The discharge efficiency of the battery in the i-th battery energy storage unit is η1, so the objective function model of the battery loss cost in the battery energy storage unit is established as: Where P i is the output power of the i-th battery energy storage unit, T is the working time of the battery energy storage unit, and n is the number of battery energy storage units.

4. According to claim 1, a method for optimizing power allocation of a large energy storage power station considering dynamic reconfigurable battery output constraints is characterized in that: The specific method of step S3 is as follows: S3.1: Consider that the heat generation power is related to the output power, where the heat generation power includes the heat generation power of the battery and the heat generation power of the inverter; Q A (i)=Q B (i)+Q C (i) (4) In the formula, Q A (i) is the total heat generation power of the battery energy storage unit, Q B (i) is the heat generation power of the battery, Q C (i) is the heat generation power of the inverter; Where P dis is the discharge power of the battery energy storage unit, η1 is the discharge efficiency of the energy storage battery, and η2 is the output efficiency of the inverter; S3.2: Assume that the output power is constant. Considering that n battery energy storage units work for t hours, the objective function model of the temperature regulation cost is established as: In the formula, K is the ratio of heat generation or heat removal power to electrical power, P pri is the electricity purchase price, C A (i) The cost required for temperature regulation.

5. According to claim 1, a method for optimizing power allocation of a large energy storage power station considering dynamic reconfigurable battery output constraints is characterized in that: The specific method of step S4 is as follows: S4.1: Consider the state of charge of the energy storage battery after power output, and its update formula is: SOC new =SOC i -((P i *TEA max ) (7) In the formula, SOC i is the state of charge of the battery of the i-th battery energy storage unit, E max is the maximum capacity of the energy storage battery, T is the working time of the battery energy storage unit, SOC new It is the state of charge of the energy storage battery after power output; S4.2: Establish an objective function model considering the consistency of the state of charge parameters of the battery energy storage unit: In the formula, SOC avg It is the average value of the state of charge of the energy storage battery after power output.

6. A method for optimizing power allocation of a large energy storage power station considering dynamic reconfigurable battery output constraints according to claim 1, characterized in that: The specific method of step S5 is as follows: S5.1: Refer to the fuzzy set theory and use the membership function F to describe the optimization results of each objective function. The membership function F is defined as follows: In the formula, F i (X) is the target f i The membership function of (X), X i * is the optimal strategy for target optimization, f iw It is the worst value of the target when optimizing each single target; S5.2: Use linear weighting to transform the multi-objective into a single-objective optimization. The objective function of the single-objective optimization is: In the formula, w i is the weight coefficient; S5.3: Use the multivariate comparison weighting method to determine the fuzzy scale of each target, convert the fuzzy mood operator into the corresponding membership, and normalize the membership to obtain the weight coefficient w i .

7. A method for optimizing power allocation of a large energy storage power station considering dynamic reconfigurable battery output constraints according to claim 6, characterized in that: Weight coefficient w i Satisfies the following relationship:

8. A method for optimizing power allocation of a large energy storage power station considering dynamic reconfigurable battery output constraints according to claim 6, characterized in that: Use the comparison and weight determination method of qualitative ranking and quantitative scaling to determine the weight coefficient. The steps are as follows: For an index set consisting of n indicators {F1, F2, …, F n }, firstly, the importance of each indicator is compared in binary. If the indicator F k Than F l More important, then set the sorting scale e kl =1,e lk =0; otherwise, if F l More important, set e kl =0,e lk =1; if F k and F l Equally important, set e kl =e lk =0.5; Based on these comparison results, the importance qualitative ranking scale matrix E is constructed: And=(and kl ) n×n (12) The matrix E needs to be subjected to a consistency test to ensure that there is no logical contradiction. The specific rules are as follows: If e hk >e hl , then there should be e lk >e kl ; If e hk <e hl , then there should be e lk <e kl ; If e hk =e hl =0.5, then e lk =e kl =0.5; Then, the sum of the elements in each row of the matrix E is calculated, and the qualitative ranking of the indicator set is determined according to the order of their size; by comparing different indicators, the importance of each indicator is described using fuzzy language and text, and then these fuzzy tones are converted into corresponding membership values, and normalized to obtain the weight coefficient w of the indicator i .

9. A method for optimizing power allocation of a large energy storage power station considering dynamic reconfigurable battery output constraints according to claim 1, characterized in that: In step S7, the particle swarm algorithm based on the fuzzy set theory is used to solve the multi-objective function model. The specific steps are as follows: ① Solve the optimal solution and the worst solution of each designed single-objective function model; ② Initialize the particle swarm, including the swarm size N, the position x of each particle i and speed v i ; ③ Use the multiple comparison and weight determination method to determine the weight coefficient; ④ Calculate the membership degree of each particle and perform normalization; ⑤ Calculate the fuzzy objective function value Fit(i) of each particle. The formula is as follows: ⑥ For each particle, compare its fuzzy objective function value Fit(i) with the individual extreme value pbest(i). If Fit(i) < pbest(i), then update the value of pbest(i); ⑦For each particle, compare its fuzzy objective function value Fit(i) with the global extreme value gbest. If Fit(i) < gbest, then update the value of gbest; ⑧Iteratively update the particle's velocity v i and position x i ; ⑨Perform boundary condition processing, that is: in min ≤in i ≤in max (14); xmin ≤ xi ≤ xmax ⑩Judge whether the algorithm termination condition is satisfied. If it is satisfied, stop the operation and output the optimization result; otherwise, return to step ⑤ and repeat this process until the end.

Images