Distributed multi-element energy storage power station optimization scheduling method considering cluster division

Abstract

The invention discloses a distributed multi-element energy storage power station optimization scheduling method considering cluster division, and relates to the field of energy storage, and the method comprises the steps: calculating three indexes of electrical distance, energy storage energy efficiency and energy storage response through the parameters of a distributed multi-element energy storage power station, carrying out the normalization of the three indexes, and carrying out the weighted average of the three indexes, so as to obtain a comprehensive modularization degree Q; and dividing an energy storage cluster by using a community discovery method based on the comprehensive modularization degree Q, taking the divided cluster as a virtual energy storage unit for aggregation, and finally solving in the energy storage collaborative optimization scheduling model. The distributed multi-element energy storage power station optimal scheduling method considering cluster division is established, and the distributed energy storage clusters are divided and aggregated by considering the electrical distance and the multi-element energy storage performance characteristics, so that the purpose of fully extracting the characteristics of the energy storage clusters to perform cluster division is achieved.

Description

Technical Field

[0001] This invention relates to the field of energy storage, specifically a method for optimizing the scheduling of distributed multi-element energy storage power stations that considers cluster partitioning. Background Technology

[0002] As a key link in the energy system's emission reduction efforts, the power industry will bear greater responsibility for carbon reduction. With the large-scale integration of renewable energy sources such as wind and solar power, their output exhibits significant randomness and volatility, further exacerbating the difficulty of balancing power generation and consumption across time and space. As a high-quality, flexible regulation resource, energy storage technology possesses rapid response and bidirectional regulation capabilities, and has become a crucial means to support the friendly integration of new energy sources and enhance system flexibility and stability. In the rapidly evolving energy landscape, energy storage resources are showing a trend towards wide-area deployment, decentralized ownership, and diverse types.

[0003] However, energy storage in real-world systems is often distributed and diversified, including electrochemical energy storage, pumped hydro storage, and flywheel energy storage, each with different operating mechanisms, response characteristics, and efficiency levels. Modeling each energy storage unit independently for scheduling would drastically increase the number of variables and the dimensionality of combinations in the scheduling model, significantly increasing solution time. Therefore, power systems often use cluster partitioning and aggregation methods to handle these scenarios, explore the regulation potential of diverse novel energy storage systems, and effectively evaluate system-level scheduling performance. Summary of the Invention

[0004] The purpose of this invention is to provide an optimized scheduling method for distributed multi-element energy storage power stations that considers cluster partitioning, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] An optimized scheduling method for distributed multi-element energy storage power stations considering cluster partitioning includes the following steps:

[0007] Step 1: Establish electrical distance index, energy efficiency coefficient index, and response coefficient index. By considering these three indexes, extract the characteristics of distributed multi-element energy storage power stations for cluster partitioning.

[0008] Step 2: The electrical distance index, energy efficiency coefficient index, and response coefficient index are weighted and summed as the basis for calculating the comprehensive modularity Q. The Louvain community discovery algorithm is used to divide the distributed multi-element energy storage cluster.

[0009] Step 3: Establish an energy storage optimization scheduling model. Based on the unit combined operation system, set the system operating cost as the objective function, and consider power balance, wind power output, photovoltaic power output and energy storage equipment constraints.

[0010] Step 4: Model solution. Initialize the parameters of each unit and energy storage power station in the system, predict the output of wind power and photovoltaic power stations and the load curves of each node in the system, linearize the optimization model and solve for the optimized scheduling results.

[0011] As a further preferred embodiment of the present invention: the electrical distance reflects the mutual influence of power and voltage changes, the degree of system coupling, and the clustering characteristics of the network, and the electrical distance index is established as follows:

[0012] d ij =d ji =Z ii +Z jj -2·Z ij (20)

[0013]

[0014] In the formula: d ij d ji Z represents the electrical impedance distance between node i and node j. ii Z jj Z represents the self-impedance of nodes i and j; ij D is the mutual impedance between nodes i and j; ij ε represents the relative distance between the two nodes in the entire system; ε is a minimal quantity to avoid division by zero.

[0015] As a further preferred embodiment of the present invention, the energy efficiency coefficient index is established as follows:

[0016]

[0017] In the formula: E n,max E n,min These represent the maximum and minimum energy storage capacity connected at node n, respectively. For energy storage charging and discharging efficiency; These represent the maximum charging power and maximum discharging power of the energy storage, respectively; C n The similarity of the energy efficiency indices of two nodes, representing the levelized cost of energy storage, is measured by a Gaussian kernel function and denoted as S. 1ij ; σ1 is t n The standard deviation of the ...

[0018] As a further preferred embodiment of the present invention, the response coefficient index is established as follows:

[0019]

[0020] In the formula: R nThe power ramp rate of energy storage; ΔP n Let S be the power change at node n within a unit time period. The response coefficient reflects the energy storage system's ability to respond to short-term load fluctuations. The similarity of the energy efficiency indices of two nodes is measured by the Gaussian kernel function and is expressed as S. 2ij ; σ2 is f n The larger the standard deviation coefficient, the higher the degree of matching between the energy storage response and load changes at the corresponding node.

[0021] As a further preferred embodiment of the present invention: step 2 specifically involves using the degree of comprehensive modularity Q as an evaluation index for community division within the system.

[0022] B=w1·D'+w2·S'1+w3·S'2 (26)

[0023]

[0024] In the formula: w1, w2, and w3 are the weights of the indicators; D', S'1, and S'2 are the normalized values ​​of the corresponding indicators; m is the sum of the weights of all edges in the network; k i It is the sum of the weights of all edges connected to node n;

[0025] First, construct an interconnected network. Based on the electrical distance, geographical location, energy efficiency coefficient, and response capability indicators between energy storage devices or system nodes, construct a weighted undirected graph. In the graph, nodes represent electrical nodes, and edge weights represent the similarity or connection strength between them.

[0026] Secondly, the optimization objective is to maximize the overall modularity Q of the network partitioning. A heuristic strategy is adopted to iteratively perform two-stage operations: In the first stage, each node is initially regarded as an independent community, and the modularity is improved by moving nodes locally; In the second stage, nodes in the same community are merged into "super nodes" to build a new network, and the operation of the first stage is repeated on the new network until the modularity no longer improves.

[0027] Finally, the partitioning results are output. The community structure output by the algorithm is the partitioning result of the energy storage cluster. The energy storage units within each cluster are similar in electrical connection and operating characteristics, and participate in scheduling as a virtual unit.

[0028] As a further preferred embodiment of the present invention: the energy storage optimization scheduling model considers thermal power output and renewable energy generation, and the aggregated energy storage cluster is embedded in the scheduling process:

[0029] Objective function:

[0030] The objective function of the scheduling model is to minimize the cost function, as shown in the following equation:

[0031] min Ftotal =C G +C S +C N (30)

[0032] In the formula: F total C represents the total economic operating cost of the system. G For the operating cost of all thermal power generating units, C S For the operating cost of all energy storage devices, C N The punitive costs of cutting renewable energy;

[0033] Constraints

[0034] 1) System power balance constraints

[0035]

[0036] In the formula: where P G,i P pv,i P wt,i P is the power provided by device i. es,i It is the power of energy storage or cluster i, L n,i It is the load of node i;

[0037] 2) Wind and solar power output constraints

[0038] 0≤D W ≤1 (32)

[0039] 0≤D PV ≤1 (33)

[0040] In the formula: D W D PV The curtailment rates for wind power and solar power;

[0041] 3) Constraints of thermal power units

[0042]

[0043] u i,t P G,i,min ≤P G,i ≤u i,t P G,i,max (35)

[0044] -R G,i ≤P G,i,t -P G,i,t-1 ≤R G,i (36)

[0045]

[0046] In the formula: u i,tLet P be the state variable of generator unit i at time t. G,i,max P G,i,min R represents the maximum and minimum power output of generator set i. G,i TS and TO represent the maximum climbing rate of generator set i, and the start-up and shutdown times of generator set t, respectively.

[0047] 4) Constraints of Energy Storage Clusters

[0048] Each energy storage cluster is considered as a virtual energy storage unit, and its operational constraints are defined as follows:

[0049]

[0050] In the formula: This refers to the charging and discharging power. This is the maximum charge / discharge power; Let be the charge / discharge state parameters of the i-th energy storage cluster at time t; For charge / discharge efficiency; S i,t E represents the percentage of capacity of the i-th energy storage cluster at time t; max Rated capacity; S i,max S i,min These are the upper and lower limits for the capacity percentage.

[0051] As a further preferred embodiment of the present invention: Step 4 specifically involves: initializing the parameters of each device in the system, inputting the load curves of each node and the output of wind power and photovoltaic power, linearizing the optimization model, calling the solver to solve the model, and obtaining the collaborative optimization scheduling results of the energy storage cluster.

[0052] Compared with the prior art, the beneficial effects of the present invention are:

[0053] (1) This invention establishes an optimized scheduling method for distributed multi-element energy storage power stations that considers cluster partitioning. By considering electrical distance and the performance characteristics of multi-element energy storage (energy efficiency coefficient, response coefficient), the distributed energy storage cluster is partitioned and aggregated, thereby achieving the purpose of fully extracting the characteristics of the energy storage cluster for cluster partitioning.

[0054] (2) By aggregating the energy storage units that have completed cluster division into virtual energy storage units and embedding them into a system containing thermal power units and renewable energy for optimized scheduling analysis, the model solution time is greatly reduced and the scheduling efficiency is significantly improved. Attached Figure Description

[0055] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0056] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] This invention addresses the optimization scheduling problem of distributed multi-element energy storage systems by proposing a cluster-based optimization scheduling method. It effectively extracts the characteristics of energy storage nodes by combining indicators such as electrical distance, energy efficiency coefficient, and response coefficient, and employs the Louvain community detection algorithm to partition the energy storage devices into clusters. The partitioned energy storage clusters are then embedded as virtual energy storage units into a system containing thermal power units and renewable energy sources for optimization scheduling analysis, significantly reducing model solution time and substantially improving scheduling efficiency.

[0058] The above-mentioned technical problems of the present invention are mainly solved by the following technical solutions:

[0059] Step 1: Establish electrical distance index, energy efficiency coefficient index, and response coefficient index. By considering these three indexes, extract the characteristics of distributed multi-element energy storage power stations for cluster partitioning.

[0060] Step 2: The electrical distance index, energy efficiency coefficient index, and response coefficient index are weighted and summed as the basis for calculating the comprehensive modularity Q. The Louvain community discovery algorithm is used to divide the distributed multi-element energy storage cluster.

[0061] Step 3: Establish an optimized scheduling model. Based on the unit combined operation system, set the system operating cost as the objective function, and consider power balance, wind power output, photovoltaic power output and energy storage device constraints.

[0062] Step 4: Model solution. Initialize the parameters of each unit and energy storage power station in the system, predict the output of wind power and photovoltaic power stations and the load curves of each node in the system, linearize the optimization model and solve for the optimized scheduling results.

[0063] Distributed multi-element energy storage cluster partitioning method:

[0064] Community detection aims to identify subgroups of nodes with dense internal connections and sparse external connections, thereby revealing the modular structure of power systems with complex structures and multi-level coupling relationships. This method integrates structural information of the power system and characteristics of different types of energy storage, and introduces three modularity-based indices to support community detection.

[0065] 1) Electrical distance index

[0066] By comprehensively considering the constraints of regional power grid topology, the dispersion and organizational logic of energy storage systems, and the electrical correlation between representative nodes, an electrical distance index is proposed to describe the community structure of power networks. As a key indicator of electrical correlation between nodes, electrical distance reflects the mutual influence under power and voltage variations, the degree of system coupling, and the clustering characteristics of the network.

[0067] d ij =d ji =Z ii +Z jj -2·Z ij (39)

[0068]

[0069] In the formula: d ij d ji Z represents the electrical impedance distance between node i and node j. ii Z jj Z represents the self-impedance of nodes i and j; ij D is the mutual impedance between nodes i and j; ij ε represents the relative distance between the two nodes in the entire system; ε is a minimal quantity to avoid division by zero.

[0070] 2) Energy storage efficiency indicators

[0071] Different types of energy storage devices vary significantly in terms of power generation principles, energy conversion efficiency, available capacity, and levelized cost of energy. Therefore, it is necessary to establish a charge-discharge efficiency index that reflects these characteristics.

[0072]

[0073] In the formula: E n,max E n,min These represent the maximum and minimum energy storage capacity connected at node n, respectively. For energy storage charging and discharging efficiency; These represent the maximum charging power and maximum discharging power of the energy storage, respectively; C n This represents the levelized cost of electricity (LCOE) for energy storage. The similarity of the energy efficiency metrics between two nodes is measured by a Gaussian kernel function, denoted as S. 1ij ; σ1 is t n The standard deviation of the value. This indicator reflects the longest duration of energy utilization per unit cost. When no energy storage device is connected to the node, this indicator is set to 0. The same applies to the following indicators.

[0074] 3) Energy storage response indicators

[0075] Different types of energy storage devices exhibit significant differences in response characteristics. Electrochemical and flywheel energy storage can respond rapidly to grid regulation demands in a short period, while pumped storage, due to its longer start-up time and turbine operating limitations, requires consideration of slope constraints. Therefore, we introduce a response coefficient index to describe the energy storage system's ability to respond to short-term fluctuations in net load.

[0076]

[0077] In the formula: R n The power ramp rate of energy storage; ΔP n Let S be the power change at node n within a unit time period. This indicator reflects the energy storage system's response to short-term load fluctuations. The similarity of the energy efficiency indicators of two nodes is measured by a Gaussian kernel function, denoted as S. 2ij ; σ2 is f n The standard deviation of the coefficient indicates the degree of matching between the energy storage response and load changes at the corresponding node.

[0078] 4) Energy storage cluster partitioning method based on community discovery

[0079] The definition of modularity references the electrical distance under the constraints of system network topology and further incorporates two characteristic indicators of heterogeneous energy storage. The resulting comprehensive modularity Q can be used as an evaluation index for community division within the system.

[0080] B=w1·D'+w2·S'1+w3·S'2 (45)

[0081]

[0082]

[0083] In the formula: w1, w2, and w3 are the weights of the indicators; D', S'1, and S'2 are the normalized values ​​of the corresponding indicators; m is the sum of the weights of all edges in the network; k i It is the sum of the weights of all edges connected to node n.

[0084] First, construct the interconnected network. A weighted undirected graph is built based on indicators such as electrical distance, geographical location, energy efficiency coefficient, and response capability between energy storage devices or system nodes. Nodes in the graph represent electrical nodes, and edge weights represent the similarity or connection strength between them.

[0085] Secondly, the optimization objective is to maximize the overall modularity Q of the network partitioning. This method employs a heuristic strategy to iteratively perform two phases of operation: in the first phase, each node is initially treated as an independent community, and the modularity is improved by locally moving nodes; in the second phase, nodes in the same community are merged into "supernodes" to construct a new network, and the operation of the first phase is repeated on the new network until the modularity no longer improves.

[0086] Finally, the partitioning results are output. The community structure ultimately output by the algorithm is the partitioning result of the energy storage cluster. The energy storage units within each cluster are similar in electrical connection and operating characteristics, and can participate in scheduling as a virtual unit.

[0087] Establishment of an energy storage optimization scheduling model:

[0088] Based on the established energy storage resource clusters and aggregations, this invention establishes an energy storage optimization scheduling model that considers thermal power output and renewable energy generation, with the aggregated energy storage clusters embedded in the scheduling process.

[0089] 1. Objective function

[0090] The objective function of the scheduling model is to minimize the cost function, as shown in the following equation:

[0091] min F total =C G +C S +C N (49)

[0092] In the formula: F total C represents the total economic operating cost of the system. G For the operating cost of all thermal power generating units, C S For the operating cost of all energy storage devices, C N The penalty costs for cutting renewable energy.

[0093] 2. Constraints

[0094] 1) System power balance constraints

[0095]

[0096] In the formula: where P G,i P pv,i P wt,i P is the power provided by device i. es,i It is the power of energy storage or cluster i, L n,i It is the load of node i.

[0097] 2) Wind and solar power output constraints

[0098] 0≤D W≤1 (51)

[0099] 0≤D PV ≤1 (52)

[0100] In the formula: D W D PV This refers to the curtailment rate of wind power and solar power.

[0101] 3) Constraints of thermal power units

[0102]

[0103] u i,t P G,i,min ≤P G,i ≤u i,t P G,i,max (54)

[0104] -R G,i ≤P G,i,t -P G,i,t-1 ≤R G,i (55)

[0105]

[0106] In the formula: u i,t Let P be the state variable of generator unit i at time t. G,i,max P G,i,min R represents the maximum and minimum power output of generator set i. G,i TS represents the maximum climbing rate of generator unit i, and TS and TO represent the start-up and shutdown times of generator unit t.

[0107] 4) Constraints of Energy Storage Clusters

[0108] In the timescale of power system dispatch, different physical energy storage devices can be approximated as operating under steady-state conditions, with transient dynamics ignored. This allows for the extraction of common steady-state characteristics and the development of a unified model to represent their external operational behavior. Treating each energy storage cluster as a virtual energy storage unit, its operational constraints are defined as follows:

[0109]

[0110] In the formula: This refers to the charging and discharging power. This is the maximum charge / discharge power; Let be the charge / discharge state parameters of the i-th energy storage cluster at time t; For charge / discharge efficiency; S i,t E represents the percentage of capacity of the i-th energy storage cluster at time t; max Rated capacity; S i,max Si,min These are the upper and lower limits for the capacity percentage.

[0111] Model solution:

[0112] Initialize the parameters of each device in the system, input the load curves of each node and the output of wind power and photovoltaic power, linearize the optimization model, call the solver to solve the model, and obtain the collaborative optimization scheduling results of the energy storage cluster.

[0113] The technical problem solved by this invention is to provide an optimized scheduling method for distributed multi-element energy storage power stations that considers cluster partitioning. Specifically, it divides and aggregates distributed energy storage clusters by considering the influence of electrical distance and the performance characteristics (energy efficiency coefficient, response coefficient) of multi-element energy storage, so as to achieve reasonable energy storage cluster partitioning while ensuring the economical and efficient operation of the energy system.

[0114] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0115] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for optimal scheduling of distributed multi-element energy storage power stations considering cluster partitioning, characterized in that, Includes the following steps: Step 1: Establish electrical distance index, energy efficiency coefficient index, and response coefficient index, and extract the characteristics of distributed multi-element energy storage power stations for cluster division; Step 2: The electrical distance index, energy efficiency coefficient index, and response coefficient index are weighted and summed as the basis for calculating the comprehensive modularity Q. The Louvain community discovery algorithm is used to divide the distributed multi-element energy storage cluster. Step 3: Establish an energy storage optimization scheduling model. Based on the unit combined operation system, set the system operating cost as the objective function, and consider power balance, wind power output, photovoltaic power output and energy storage equipment constraints. Step 4: Model solution. Initialize the parameters of each unit and energy storage power station in the system, predict the output of wind power and photovoltaic power stations and the load curves of each node in the system, linearize the optimization model and solve for the optimized scheduling results.

2. The method for optimal scheduling of distributed multi-element energy storage power stations considering cluster partitioning as described in claim 1, characterized in that, The electrical distance reflects the mutual influence of power and voltage changes, the degree of system coupling, and the clustering characteristics of the network. The electrical distance index is established as follows: d ij =d ji =Z ii +Z jj -2·Z ij (1) In the formula: d ij d ji Z represents the electrical impedance distance between node i and node j. ii , Z jj Z represents the self-impedance of nodes i and j; ij D is the mutual impedance between nodes i and j; ij ε represents the relative distance between the two nodes in the entire system; ε is a minimal quantity to avoid division by zero.

3. The method for optimal scheduling of distributed multi-element energy storage power stations considering cluster partitioning as described in claim 1, characterized in that, The energy efficiency coefficient indicators are established as follows: In the formula: E n,max E n,min These represent the maximum and minimum energy storage capacity connected at node n, respectively. For energy storage charging and discharging efficiency; These represent the maximum charging power and maximum discharging power of the energy storage, respectively; C n The similarity of the energy efficiency indices of two nodes, representing the levelized cost of energy storage, is measured by a Gaussian kernel function and denoted as S. 1ij ; σ1 is t n The standard deviation of the ...

4. The method for optimal scheduling of distributed multi-element energy storage power stations considering cluster partitioning as described in claim 1, characterized in that, The response coefficient index is established as follows: In the formula: R n The power ramp rate of energy storage; ΔP n Let S be the power change at node n within a unit time period. The response coefficient reflects the energy storage system's ability to respond to short-term load fluctuations. The similarity of the energy efficiency indices of two nodes is measured by the Gaussian kernel function and is expressed as S. 2ij ; σ2 is f n The larger the standard deviation coefficient, the higher the degree of matching between the energy storage response and load changes at the corresponding node.

5. The method for optimal scheduling of distributed multi-element energy storage power stations considering cluster partitioning as described in claim 1, characterized in that, Step 2 specifically involves using the degree of comprehensive modularity Q as an evaluation index for community division within the system. B=w1·D'+w2·S'1+w3·S'2 (7) In the formula: w1, w2, and w3 are the weights of the indicators; D', S'1, and S'2 are the normalized values ​​of the corresponding indicators; m is the sum of the weights of all edges in the network; k i It is the sum of the weights of all edges connected to node n; First, construct an interconnected network. Based on the electrical distance, geographical location, energy efficiency coefficient, and response capability indicators between energy storage devices or system nodes, construct a weighted undirected graph. In the graph, nodes represent electrical nodes, and edge weights represent the similarity or connection strength between them. Secondly, the optimization objective is to maximize the overall modularity Q of the network partitioning. A heuristic strategy is adopted to iteratively perform two-stage operations: In the first stage, each node is initially regarded as an independent community, and the modularity is improved by moving nodes locally; In the second stage, nodes in the same community are merged into "super nodes" to build a new network, and the operation of the first stage is repeated on the new network until the modularity no longer improves. Finally, the partitioning results are output. The community structure output by the algorithm is the partitioning result of the energy storage cluster. The energy storage units within each cluster are similar in electrical connection and operating characteristics, and participate in scheduling as a virtual unit.

6. The method for optimal scheduling of distributed multi-element energy storage power stations considering cluster partitioning as described in claim 1, characterized in that, The energy storage optimization scheduling model considers both thermal power output and renewable energy generation, and the aggregated energy storage clusters are embedded in the scheduling process: Objective function: The objective function of the scheduling model is to minimize the cost function, as shown in the following equation: minF total =C G +C S +C N (11) In the formula: F total C represents the total economic operating cost of the system. G For the operating cost of all thermal power generating units, C S For the operating cost of all energy storage devices, C N The punitive costs of cutting renewable energy; Constraints 1) System power balance constraints In the formula: where P G,i 、P pv,i 、P wt,i P is the power provided by device i. es,i It is the power of energy storage or cluster i, L n,i It is the load of node i; 2) Wind and solar power output constraints 0≤D W ≤1 (13) 0≤D PV ≤1 (14) In the formula: D W D PV The curtailment rates for wind power and solar power; 3) Constraints of thermal power units u i,t Q G,i,min ≤P G,i ≤u i,t Q G,i,max (16) -R G,i ≤P G,i,t -P G,i,t-1 ≤R G,i (17) In the formula: u i,t Let P be the state variable of generator unit i at time t. G,i,max 、P G,i,min R represents the maximum and minimum power output of generator set i. G,i TS and TO represent the maximum ramp rate of generator set i, and TS and TO represent the start-up and shutdown times of generator set t, respectively. 4) Constraints of Energy Storage Clusters Each energy storage cluster is considered as a virtual energy storage unit, and its operational constraints are defined as follows: In the formula: This refers to the charging and discharging power. This is the maximum charge / discharge power; Let be the charge / discharge state parameters of the i-th energy storage cluster at time t; For charge / discharge efficiency; S i,t E represents the percentage of capacity of the i-th energy storage cluster at time t; max Rated capacity; S i,max 、S i,min These are the upper and lower limits for the capacity percentage.

7. The method for optimal scheduling of distributed multi-element energy storage power stations considering cluster partitioning as described in claim 1, characterized in that, Step 4 specifically involves: initializing the parameters of each device in the system, inputting the load curves of each node and the output of wind power and photovoltaic power, linearizing the optimization model, calling the solver to solve the model, and obtaining the collaborative optimization scheduling results of the energy storage cluster.

Images