Optimal configuration method and system for distributed energy storage system of power distribution network

Abstract

The invention discloses an optimal configuration method and system for a distributed energy storage system of a power distribution network. The method comprises the following steps: acquiring state information of each power distribution network node in a target area; calculating the carbon emission of each power distribution network node according to the state information of each power distribution network node, calculating the influence parameter of cluster energy storage on the carbon emission reduction of each power distribution network node, and carrying out cluster partitioning on the planning position of energy storage so as to screen out an energy storage system installation area in advance; the method comprises the following steps: establishing a calculation model of the comprehensive vulnerability of a line, identifying the position of a weak node of a target energy storage system, establishing a target optimization model of energy storage locating and sizing by taking the lowest operation cost and the optimal comprehensive vulnerability of the line as targets, and solving the model to obtain the optimal installation position and the optimal capacity configuration of the energy storage system. According to the method, the system operation cost, the carbon emission influence and the system vulnerability can be comprehensively considered, the optimal energy storage system configuration scheme is determined, the operation cost and the carbon emission are reduced, and meanwhile the toughness of the power distribution network is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular, to an optimal configuration method and system for a distribution network distributed energy storage system. Background Art

[0002] Battery Energy Storage Systems (BESSs) have fast power regulation capabilities. They can not only effectively suppress the fluctuations of photovoltaic power output, but also improve voltage quality, reduce the peak-valley difference, etc. Energy storage shows a multi-point distribution characteristic in the power grid. Through the unified dispatching of the power system, the multi-point layout energy storage system can realize the orderly aggregation of multi-point distributed energy storage. In addition to meeting the local application function, it can also provide emergency power support for the power grid, improve the safety and stability of the power grid, effectively enhance the power grid's ability to absorb renewable energy, and enrich power grid peak shaving, frequency modulation, and voltage regulation, etc., making the power system more "flexible" and "intelligent". Therefore, the selection of the access location, installed power, and capacity of the battery energy storage system will directly affect the efficient and economic operation of the distribution network.

[0003] Currently, for the configuration of distributed energy storage systems, the installation location selection and capacity configuration usually only consider improving the power quality of the distribution network or reducing voltage fluctuations. However, the line structure vulnerability and line state vulnerability of different systems are different, and the difficulty of transmitting electric energy between nodes in the line, the impact of the line on network power changes, and the probability of power quality problems in the line are also different. Therefore, there are weak nodes in the system that can cause a decline in network connectivity, transmission capacity, etc. The energy storage system can ensure the power supply of important loads and improve the reliability of the system. At this time, if an extreme disaster occurs in the system, the traditional configuration methods according to power quality and economy cannot reinforce the weak nodes in the system, affecting the reliability of the system and resulting in the system's inability to face disasters. Summary of the Invention

[0004] The technical problem to be solved by the present invention lies in: aiming at the technical problems existing in the prior art, the present invention provides an optimal configuration method and system for a distribution network distributed energy storage system with a high implementation method, low cost, and high configuration accuracy, which can comprehensively consider the system operation cost, carbon emission impact, and system vulnerability improvement to determine the optimal configuration plan of the energy storage system.

[0005] To solve the above technical problems, the technical solution proposed by the present invention is as follows:

[0006] An optimal configuration method for a distribution network distributed energy storage system, the steps include:

[0007] Obtain the status information of each distribution network node in the target area, where the status information includes node connection status, power output data of the power source, and load data;

[0008] Calculate the carbon emissions of each distribution network node according to the state information of each distribution network node, calculate the influence parameters of the cluster energy storage on the carbon emission reduction of each distribution network node according to the carbon emissions of each distribution network node, and perform cluster partitioning on the planned location of the energy storage according to the influence parameters to pre-screen the installation area of the energy storage system;

[0009] Use a variety of vulnerability characteristics used to characterize the line vulnerability characteristics of the energy storage system to establish a calculation model for the comprehensive line vulnerability through weighting, so as to identify the weak node positions of the target energy storage system. With the lowest operating cost and the optimal comprehensive line vulnerability as the goals, establish an objective optimization model for energy storage site selection and capacity determination, and input the state information of each distribution network node in the target area to solve the objective optimization model to obtain the optimal installation position within the installation area of the energy storage system and the optimal capacity configuration of the energy storage system.

[0010] Further, calculate the carbon emissions of the distribution network node according to the following formula:

[0011]

[0012] Among them, P l is the active power flowing into node i of the branch; ρ l is the carbon flow density of branch l; is the set of all branches with active power inflow of the power flow; G i is the equivalent unit injection power of the i-th node; P bi is the sum of the output powers of the energy storage and new energy; P N is the node active power flux matrix; P B is the branch power flow distribution matrix; P G is the injection distribution matrix of the energy storage and new energy units. When the equivalent load P le is less than 0, it corresponds to the absolute value of P bi equal to P le ; E N is the node carbon potential column vector, indicating the carbon potential level of each node, and R L represents the node load carbon flow rate; EG is an N-order column vector composed of the equivalent unit carbon emission intensity e gi ;

[0013] Further, calculate the influence parameters of the cluster energy storage on the carbon emission reduction of each distribution network node according to the following formula:

[0014]

[0015] Among them, ΔR ij represents the influence parameter of the node carbon emissions flowing from node i to node j on the carbon emissions of the power grid, and ΔP lis the contribution of the cluster energy storage to the load power of this node; e l is an N-1 order unit transverse vector, e li is an N-1 order transverse vector with the i-th bit being 1 and the rest being 0; E N is a column vector composed of the carbon potentials of each node before being affected by the cluster energy storage;

[0016] Give priority to ΔR ij Supply power to the nodes whose values are less than the preset threshold, and deliver the surplus to other nodes in the cluster, and perform cluster partitioning on the planned positions of the energy storage according to the ΔR ij value.

[0017] Furthermore, the step of performing cluster partitioning on the planned positions of the energy storage according to the ΔR ij value includes:

[0018] Calculate the ΔR ij values between all nodes to construct an influence matrix, and construct a weighted graph model according to the influence matrix and network characteristics, where the nodes represent the equipment in the power grid, and the edge weights are ΔR ij values to represent the degree of carbon emission influence between nodes;

[0019] Partition each node using a community detection algorithm or a clustering algorithm to gather the nodes with similar carbon emission influences in one partition;

[0020] Determine the installation area of the preliminarily planned energy storage system according to the partition result.

[0021] Furthermore, the vulnerability features include system structure vulnerability features for characterizing the vulnerability of the system structure and system state vulnerability features for characterizing the vulnerability of the system state. The system structure vulnerability features are constructed using the line electrical betweenness in the system. The system state vulnerability features include a system power level coefficient for characterizing the system power level state, a system shock transfer ratio for characterizing the influence of line disconnections in the system on the power distribution of the remaining lines in the network, and a system voltage deviation rate for characterizing the deviation of the voltage level of each line in the system from the reference rated voltage level.

[0022] Furthermore, the calculation expression of the system structure vulnerability features is:

[0023]

[0024] where f B is the system line structure vulnerability feature, B ave is the average value of the line electrical betweenness in the system after the distributed power source is connected to the system, B ave.max is the maximum value of B ave ; m is the number of lines; l is the set of lines; B iis the electrical betweenness of line i; B e (m,n) is the electrical betweenness value of line (m,n), where m and n are the two endpoints of the line, and G and D are the sets of power supply and load nodes respectively; P g and P d are the active powers of power supply node g and load node d respectively; Y gd is the equivalent admittance between node g and node d; N g and N d are the numbers of power supply and load nodes respectively; I gd (m,n) represents the current value induced on line (m,n) after applying a unit current between the power supply-load node pair (g,d);

[0025] The calculation expression of the system power level coefficient is:

[0026]

[0027] where, f L is the system line power level coefficient; L ave is the average value of the line power level coefficients in the system after the distributed power source is connected to the system, L ave.max is L ave 's maximum value; L i is the power level coefficient of line i; P i is the active power currently transmitted by line i; P imax is the limit transmission power of line i; m is the number of lines; l is the set of lines;

[0028] The calculation expression of the system shock transfer ratio is:

[0029]

[0030] where, f G is the system shock transfer ratio, G ave is the average value of the line power level coefficients in the system after the distributed power source is connected to the system, G ave.max is G ave 's maximum value; m is the number of lines; l is the set of lines; G i is the shock transfer ratio coefficient of line i; G i is the shock transfer ratio of line i; ΔP j is the power change amount of line j caused by the disconnection of line i; P j is the original power of line j before the disconnection of line i; N L is the total number of lines;

[0031] The calculation expression of the system voltage offset rate is:

[0032]

[0033] Among them, f DU is the system line voltage deviation rate; ΔU ave is the average value of the line voltage deviation rate in the system after the distributed power source is connected to the system, and ΔU ave.max is for ΔU ave the maximum value; m is the number of lines; l is the line set; ΔU i is the voltage deviation rate of line i, and ΔU% is the voltage deviation rate of the line; U i , U j are the voltages of the two ends of a certain line node respectively; U cr is the reference rated voltage; ΔU lim is the maximum allowable voltage deviation.

[0034] Furthermore, the established calculation model of the comprehensive resilience of the line is:

[0035] f2 = w1f B + w2f L + w3f G + w4f ΔU

[0036] Among them, f2 represents the system structure vulnerability characteristic, and f L , f G and f △U represent the system power level coefficient, the system shock transfer ratio, and the system voltage deviation rate respectively, and w1 to w4 are weight coefficients;

[0037] The calculation model of the operating cost is:

[0038] f1 = C Ess = C inv + C rep + C ps + C om + C scr - C res - C sub - C rst

[0039] Among them, C Ess is the total cost, C inv is the initial cost, C rep is the renewal and replacement cost, C ps is the charge and discharge cost, C om is the system operation and maintenance cost, C scr is the scrap disposal cost, C res is the recovery salvage value, C sub is the subsidy allowance, C rst is the comprehensive benefit of delaying grid upgrade;

[0040] The established target optimization model is: min[f1, f2].

[0041] Furthermore, it also includes setting any one or more of power balance constraints, power flow constraints, voltage constraints, charge and discharge constraints of the battery system, battery SOC constraints, and constraints on the installation location and number of batteries. The power balance constraint is:

[0042]

[0043] where n pv are respectively the installation numbers of photovoltaics; P pv,m (t) are respectively the active power outputs of the m-th photovoltaic at time t; P load (t) is the total load at time t;

[0044] The power flow constraint is:

[0045]

[0046] In the formula, P i (t), Q i (t) are respectively the active and reactive powers injected into the node, and V i (t) is the voltage of node i at time t;

[0047] The voltage constraint is:

[0048]

[0049] In the formula, and are respectively the upper and lower limits of the voltage of node i;

[0050] The constraints on the total active and reactive power limits injected by the superior power grid into the distribution network are:

[0051]

[0052] where respectively represent the upper and lower limits of the active and reactive powers of the tie line;

[0053] The charge and discharge constraints of the battery system are:

[0054]

[0055] The battery SOC constraint is:

[0056] SOC min ≤SOC i (t)≤SOC max

[0057] The constraints on the installation location and number of batteries are:

[0058] L BESS，i ∈ N nodes and L BESS，i ≠ L grid

[0059] N BESS，min ≤ N BESS，i ≤ N max，max 。

[0060] Furthermore, an improved particle swarm optimization algorithm is used to solve the target optimization model, and the steps include:

[0061] Initialize the particle population;

[0062] Randomly generate the particle population and calculate the optimal fitness and global fitness of the population;

[0063] According to the status information of each distribution network node in the target area, calculate and update the velocity and position information of the particle population;

[0064] Based on the target optimization model, calculate the fitness of the objective function and determine whether the constraint conditions are satisfied. If not, return to execute the update of the population velocity and position information and recalculate the optimal fitness of the population; if the constraint conditions are satisfied, select the individual extreme value and the global extreme value through the fitness of the particle. If the fitness of the target particle is better than the global optimal fitness, use the best fitness value of the target particle to update the individual extreme value and the global extreme value of the particle;

[0065] Determine whether the maximum preset iteration condition is reached. If so, output the final solution result; otherwise, return to execute the calculation and update the velocity and position information of the particle population;

[0066] Select the optimal solution according to the solution result, and determine the optimal installation position and optimal capacity configuration of the energy storage system.

[0067] A distributed energy storage system optimization configuration system for a distribution network includes a processor and a memory. The memory is used to store a computer program, and the processor is used to execute the computer program to execute the method as described above.

[0068] Compared with the prior art, the advantages of the present invention are as follows: By calculating the contribution of each node to carbon emissions based on the distribution of carbon emission flow in the power grid, considering the impact of cluster energy storage on carbon emission reduction of each distribution network node, clustering and zoning the planned locations of energy storage, pre-screening the installation areas of energy storage systems, then establishing a comprehensive vulnerability model from two aspects of the line structure and state of the power grid system to quantify the vulnerability performance of the distribution network. At the same time, with the low-carbon operation of the distribution network energy storage system, the improvement of comprehensive vulnerability and the operation cost as the optimization objectives, the optimal installation location and optimal capacity configuration of the energy storage system are determined, so that the system can be effectively strengthened by identifying key nodes before a disaster occurs, and considering the impact of node carbon emissions, realizing the energy storage optimization configuration that can meet the carbon emission reduction constraint and improve the resilience of the distribution network, giving full play to the performance of the distributed energy storage system, effectively improving the resilience of the system, and at the same time reducing the overall operation cost and carbon emissions of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 FIG. is a schematic diagram of the implementation process of the optimization configuration method for the distributed energy storage system of the distribution network in this embodiment.

[0070] Figure 2 FIG. is a detailed process schematic diagram for implementing the optimization configuration of the distributed energy storage system of the distribution network in a specific application embodiment of the present invention.

[0071] Figure 3 FIG. is a schematic diagram of the topological structure for simulation analysis using a typical IEEE33-node distribution network system in a specific application embodiment of the present invention.

[0072] Figure 4 FIG. is a schematic diagram of the detailed steps for implementing the solution of the target optimization model using an improved particle swarm algorithm in this embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0073] The present invention will be further described below in conjunction with the accompanying drawings of the specification and specific preferred embodiments, but the protection scope of the present invention is not limited thereby.

[0074] The vulnerability of the line structure means that when the system is subjected to disturbances or failures, the power grid has a tendency not to maintain the integrity of its topological structure and normal operation. For example, when the equivalent admittance between nodes is large, the electrical distance between nodes is very small, and it is relatively easy to transmit electric energy between the corresponding nodes. Conversely, it is more difficult to transmit electric energy. The vulnerability of the line state refers to the characteristic of abnormal changes in electrical parameters representing the operating state of the system when a fault occurs in the system. For example, when the safety threshold of power fluctuation is smaller, the vulnerability of the line is higher. When a certain line is disconnected, it will cause the power of the entire network to be redistributed, that is, the disconnection of the line will affect the power change of the network. Another example is that the smaller the voltage deviation rate of the line, the farther the line is from the voltage limit deviation level, and the greater the margin of the line voltage level from the reactive power imbalance problem in the network. At this time, the line is safer.

[0075] In the present invention, first, the carbon emissions of each distribution network node are calculated according to the status information of each distribution network node. According to the distribution of carbon emission flow in the power grid, the contribution of each node to carbon emissions is calculated. Furthermore, the influence parameters of the cluster energy storage on the carbon emission reduction of each distribution network node are calculated, the planned locations of the energy storage are clustered and partitioned, and the installation areas of the energy storage system are pre-screened. Then, a comprehensive vulnerability model is established from two aspects of the line structure and state of the power grid system to quantify the vulnerability performance of the distribution network. Furthermore, an objective optimization model for the siting and sizing of the energy storage system is established with the low-carbon operation of the distribution network energy storage system, the improvement of comprehensive vulnerability, and the operation cost as the optimization objectives. By solving the model, the optimal installation location and the optimal capacity configuration of the energy storage system are determined, so that the system can be effectively strengthened before a disaster occurs by identifying key nodes, and at the same time, considering the influence of node carbon emissions, the optimal configuration of the energy storage is realized under the constraints of carbon emission reduction, reduction of operation cost, and improvement of the resilience of the distribution network, giving full play to the performance of the distributed energy storage system, effectively improving the resilience of the system, and at the same time reducing the overall operation cost and carbon emissions of the system.

[0076] As Figure 1 shown, the steps of the method for optimizing the configuration of the distribution network distributed energy storage system in this embodiment include:

[0077] Step S01. Obtain the status information of each distribution network node in the target area, where the status information includes node connection status, power generation output data, and load data.

[0078] Specifically, the power generation output data includes photovoltaic output data, generator output data, etc. Further, information such as the node type of each node, line reactance data, reactance of the photovoltaic power connection line, system node voltage, and branch current limit can also be obtained according to actual requirements.

[0079] Considering the randomness of photovoltaic and load output, clustering algorithms such as K-means can be further used to analyze the collected photovoltaic and load historical data to obtain typical daily curves of load and photovoltaic respectively, which serve as model input data (photovoltaic, load) under various scenarios.

[0080] Step S02. Calculate the carbon emissions of each distribution network node based on the status information of each distribution network node, and calculate the impact parameters of cluster energy storage on carbon emission reduction of each distribution network node based on the carbon emissions of each distribution network node. Cluster partition the planned location of energy storage based on the impact parameters to pre-screen the energy storage system installation area.

[0081] In order to achieve cluster partitioning, this embodiment first collects the carbon emission reduction data and grid topology information of each node, and standardizes the data; then, a weighted graph model is constructed in combination with the carbon emission reduction data and network characteristics, in which the nodes represent the equipment in the grid, and the edge weights can be set based on electrical characteristics or physical distances; clustering algorithms (such as K-Means or spectral clustering) are used for partitioning to ensure that the carbon emission reduction characteristics of the nodes in the partition are similar, and then the partition results are adjusted in combination with grid stability or carbon emission reduction optimization goals, and finally the areas with similar carbon emission reduction are distinguished to pre-screen the energy storage system installation area. Optionally, the partitioning effect can also be verified by indicators (such as carbon emission reduction consistency and load balancing within the cluster), and the partitioning results can be visualized to analyze and optimize grid operation.

[0082] This embodiment first calculates the carbon potential distribution of nodes in the distribution network, that is, the distribution of carbon emission flows in the power grid, calculates the contribution of each node to carbon emissions based on the node carbon potential distribution, and then calculates the impact parameters of group energy storage on carbon emission reduction of each distribution network node as the node carbon emission potential indicator. This can effectively characterize the matching degree between the load and new energy and the carbon emission reduction capacity when absorbing new energy, so that the planned location of energy storage can be clustered and the energy storage system installation area can be pre-screened.

[0083] Specifically, the node carbon potential can be defined as the sum of the carbon flow rate flowing into the upstream branch connected to the node and the carbon flow rate injected by the generator set at this node divided by the sum of the node injection power. For pure load nodes, the carbon potential of the node has nothing to do with the load size and other working conditions of the node itself, while for other nodes, the carbon potential of the node is not only related to the carbon emission intensity of the node unit, but also to the upstream trend. Therefore, when new energy or energy storage injects active power as a zero-carbon unit, for the node connected to new energy or energy storage, when the node carbon potential injection power increases to greater than or equal to the load power, the node carbon potential must not be equal to 0, but can only be infinitely close to 0. However, the node carbon flow rate is equal to the load power multiplied by the node carbon potential, but the node carbon flow rate obtained using the traditional calculation method is not 0.

[0084] As an alternative implementation, the carbon potential distribution of nodes in the distribution network can be calculated according to the following formula:

[0085]

[0086] where E N is the column vector of node carbon potential, representing the carbon potential level of each node, and R L represents the node load carbon flow rate; E G is the N-order column vector composed of the equivalent unit carbon emission intensity e gi , P N is the node active power flux matrix; P B is the branch power flow distribution matrix; P G is the injection distribution matrix of energy storage and new energy units.

[0087] Furthermore, the carbon emissions of nodes in the distribution network are calculated according to the following formula:

[0088]

[0089] where P l is the active power flowing into node i of the branch; ρ l is the carbon flow density of branch l; is the set of all branches with active power flowing in; G i is the equivalent unit injection power of the i-th node; P bi is the sum of the output powers of energy storage and new energy; when the equivalent load P le is less than 0, corresponding to P bi equal to the absolute value of P le .

[0090] By calculating the power flow distribution factor, the relationship between the active power flow out of one node and the active power flow into another node can be characterized. At the same time, considering the proportional sharing principle in power flow tracing, it can be obtained that the contribution ratio of the carbon flow injection of the unit to the load carbon flow rate is equal to the contribution ratio to the total carbon flow rate flowing into this node. As an alternative implementation, the influence parameter of the cluster energy storage on the carbon emission reduction of each distribution network node can be calculated according to the following formula:

[0091]

[0092] where ΔR ij represents the influence parameter of the node carbon emissions flowing from node i to node j on the grid carbon emissions, ΔP l is the contribution amount of the cluster energy storage to the load power of this node; e l is the N-1 order unit horizontal vector, and e li is the N-1 order horizontal vector with 1 at the i-th position and 0 for the rest, which is used to calculate ΔR ij, which means that when \(i = j\), directly consider the contribution of the change in the load power of node \(i\) itself to the carbon emissions of the power grid. When \(i\neq j\): calculate the indirect impact of node \(i\) on node \(j\) through the power grid topology (such as power flow, node carbon potential, etc.); \(E\) N is a column vector composed of the carbon potentials of each node before being affected by the cluster energy storage.

[0093] After calculating \(\Delta R\) ij , the carbon emission contributions of the power flows between different nodes can be determined. The principle of giving priority to low-carbon can be adopted to determine the priority power supply nodes, that is, give priority to supplying power to node \(j\) with a lower \(\Delta R\) ij (less than the preset threshold), and transfer the surplus to other nodes in the cluster to reduce the overall carbon emissions.

[0094] This embodiment performs cluster partitioning based on the \(\Delta R\) ij value. The specific steps include:

[0095] Step S201. Construct a weighted graph: Calculate the \(\Delta R\) ij values between all nodes to construct an influence matrix, that is, the influence matrix is composed of the \(\Delta R\) ij values between each node and is used to represent the relationship and interaction strength between each node. According to the influence matrix and network characteristics, construct a weighted graph model, where the nodes represent the equipment in the power grid and the edge weights are the \(\Delta R\) ij values to represent the degree of carbon emission influence between nodes;

[0096] Step S202. Clustering: Use a community detection algorithm (such as the Louvain method) or a clustering algorithm (such as K-Means) to partition each node so as to cluster the nodes with similar carbon emission influences into one partition;

[0097] Step S204. Determine the installation areas of the preliminarily planned energy storage systems (energy storage nodes and transmission lines) according to the partitioning results, that is, pre-screen each partition as the installation area of the energy storage system.

[0098] The Louvain method is a community partitioning algorithm that can perform network partitioning based on the node association degree (influence weight). In a specific application embodiment, when using the Louvain method to partition each node, the Louvain method can be used to maximize the modularity to calculate the best partition, that is, regard each node as a separate community and gradually merge the communities to make the modularity \(Q\) the largest. For example, the calculation expression of the modularity \(Q\) can be:

[0099]

[0100] where \(m\) is the total edge weight, \(k_i\) is the degree of node \(i\), \(\delta(c i , \(c j) is an indicator function, E i，j represents the physical distance between nodes, which can be calculated according to the node distance matrix d i，j That is, E i，j = 1 / d i，j .

[0101] In this embodiment, by considering the impact of new energy and energy storage access on node carbon emission reduction, the carbon emissions of power grid nodes are calculated, and the impact of new energy and energy storage access on node carbon emission reduction is analyzed. A model for the impact of cluster energy storage on the carbon flow rate of the distribution network is constructed to calculate the carbon emission reduction potential of each node. Furthermore, according to the carbon emission potential of each node, the planned location of the energy storage is clustered and partitioned, and the energy storage installation area can be pre-screened, so that the carbon emission reduction potential of each node can be fully exploited.

[0102] It can be understood that when planning the energy storage installation area within each partition, one energy storage system can be planned to be installed in one area. Of course, two or more energy storage systems can also be installed in one area according to actual needs.

[0103] Step S03. Use a variety of vulnerability characteristics for characterizing the line vulnerability characteristics of the energy storage system to establish a calculation model for the comprehensive line vulnerability, which is used to identify the weak node positions of the target energy storage system. With the lowest operating cost and the optimal comprehensive line vulnerability as the goals, an objective optimization model for energy storage site selection and capacity determination is established, and the state information of each distribution network node in the target area is input to solve the objective optimization model, and the optimal installation position within the energy storage system installation area and the optimal capacity configuration of the energy storage system are obtained.

[0104] In this embodiment, a calculation model for comprehensive line vulnerability is established by considering the impacts of the structure and state of the power grid. The vulnerability characteristics specifically include system structure vulnerability characteristics for characterizing system structure vulnerability and system state vulnerability characteristics for characterizing system state vulnerability. The system structure vulnerability characteristics are constructed using the line electrical betweenness in the system. The system state vulnerability characteristics include a system power level coefficient for characterizing the system power level state, a system shock transfer ratio for characterizing the impact of line disconnection in the system on the power distribution of the remaining lines in the network, and a system voltage deviation rate for characterizing the deviation of the voltage level of each line in the system from the reference rated voltage level. Combining the line electrical betweenness, power level coefficient, shock transfer ratio, and voltage deviation rate of the system fully characterizes the vulnerability of the distribution network.

[0105] As an optional implementation manner, the system structure vulnerability characteristics can be calculated as follows:

[0106]

[0107] Among them, f B is the system line structure vulnerability characteristic, Bave is the average value of the line electrical betweenness in the system after the distributed power source is connected to the system, B ave.max is B ave The maximum value, the maximum value that appears in the single-objective function optimization during the process of connecting the distributed power source to the system, and this value is used for the normalization of the single-objective function of the system line structure vulnerability; m is the number of lines; l is the line set; B i is the electrical betweenness of line i; B e (m,n) is the electrical betweenness value of line (m,n), where m and n are the two endpoints of the line respectively, and G and D are the sets of power source and load nodes; P g and P d are the active powers of power source node g and load node d respectively; Y gd is the equivalent admittance between node g and node d; N g and N d are the numbers of power source and load nodes respectively; I gd (m,n) represents the current value induced on line (m,n) after applying a unit current between the power source-load node pair (g,d).

[0108] Y gd can measure the effective amount transmitted between node g and node d. When Y gd is very large, it means that the electrical distance between nodes is very small, then it is relatively easy to transmit electric energy between nodes. At this time, the contribution rate of this power source-load node pair to the electrical betweenness of the lines in the network is very large; when Y gd is very small, it means that the electrical distance between nodes is very large, and it is very difficult to transmit electric energy between nodes. At this time, the contribution rate of this power source-load node pair to the electrical betweenness of the lines in the network is very small.

[0109] As an optional implementation manner, taking the average value of the electrical betweenness of all lines to characterize the overall structural vulnerability of the system, the system power level coefficient can be calculated by the following calculation expression:

[0110]

[0111] where, f L is the system line power level coefficient; L ave is the average value of the line power level coefficient in the system after the distributed power source is connected to the system, L ave.max is L ave The maximum value; L i is the power level coefficient of line i; P i is the active power currently transmitted by line i; P imax is the limit transmission power of line i; m is the number of lines; l is the line set.

[0112] According to the above formula, the utilization of line capacity under normal operation conditions of the line can be quantitatively described. If the line power level coefficient is larger, it indicates that the line is closer to the limit transmission power, the safety threshold for power fluctuations is smaller, and the vulnerability of the line is higher.

[0113] The impact transfer ratio can describe the impact of line disconnection on the power distribution of the remaining lines in the network. As an alternative implementation, the calculation expression of the impact transfer ratio of the line is:

[0114]

[0115] Among them, G i is the impact transfer ratio coefficient of line i; G i is the impact transfer ratio of line i; ΔP j is the power change amount of line j caused by the disconnection of line i; P j is the original power of line j before the disconnection of line i; N L is the total number of lines.

[0116] When a certain line is disconnected, it causes the power of the entire network to be redistributed. Summing up the power change amounts of all non-disconnected lines during this process, therefore, this value can be used to describe the impact of the disconnection of a certain line on the network power change. The impact of line disconnection on the network power change is an influencing factor characterizing the vulnerability of the line. Based on the impact transfer ratio of the line, the calculation expression of the system impact transfer ratio is:

[0117]

[0118] Among them, f G is the system impact transfer ratio, G ave is the average value of the line power level coefficients in the system after the distributed power source is connected to the system, G ave.max is the maximum value of G ave the maximum value that appears in the single-objective function optimization during the process of connecting the distributed power source to the system, and this value is used for the normalization of the single-objective function of the system line power level coefficient; m is the number of lines; l is the line set.

[0119] The voltage deviation rate can represent the level of the line voltage deviating from the reference rated voltage in per-unit value. The smaller the voltage deviation rate, the farther the line is from the voltage limit deviation level, the greater the margin of the line voltage level from the reactive power imbalance problem in the network, and the safer the line is at this time. As an alternative implementation, the calculation expression of the voltage deviation rate of the line is:

[0120]

[0121] Among them, ΔU% is the voltage deviation rate of the line; Ui , U j are the voltages at the two ends of a certain line respectively; U cr is the reference rated voltage. For example, the reference rated voltage can be set to 1; ΔU lim is the maximum allowable voltage deviation.

[0122] Based on the voltage deviation rate of the line, the calculation expression of the system voltage deviation rate is:

[0123]

[0124] Among them, f DU is the system line voltage deviation rate; △U ave is the average value of the line voltage deviation rate in the system after the distributed power source is connected to the system, and △U ave.max is the maximum value of △U ave ; m is the number of lines; l is the set of lines; △U i is the voltage deviation rate of line i.

[0125] Furthermore, based on the above system line electrical betweenness, power level coefficient, impact transfer ratio, and voltage deviation rate, a calculation model for the comprehensive resilience of the line is established. To determine the weights of each index, a method combining the analytic hierarchy process and the entropy weight method can be used to comprehensively determine the weights of each index, which can reduce the error caused by subjective factor scoring in the analytic hierarchy process.

[0126] B i = W1B s + W2B r

[0127] B r = β1L i + β2G i + β3ΔU%

[0128] Among them: B i represents the comprehensive vulnerability of the line; B s and B r represent the line structure and state, and W1~W3, β1~β3 represent the weight coefficients.

[0129] As an alternative implementation, the calculation model of the comprehensive resilience of the line can be established as follows:

[0130] f2 = w1f B + w2f L + w3f G + w4f ΔU (13)

[0131] Among them, f2 represents the system structure vulnerability feature, f L , fG and f △U represent the system power level coefficient, the system shock transfer ratio, and the system voltage deviation rate respectively, and w1 to w4 are weight coefficients.

[0132] As an alternative implementation, the AHP (Analytic Hierarchy Process)-entropy weight method can be used to comprehensively determine the weights of the above indicators to identify the weak node positions of the system.

[0133] In this embodiment, a calculation model for the operating cost is also constructed. For example, the following expression can be used:

[0134] f1 = C Ess = C inv + C rep + C ps + C om + C scr - C res - C sub - C rst (14)

[0135] where C Ess is the total cost, C inv is the initial cost, C rep is the update and replacement cost, C ps is the charge and discharge cost, C om is the system operation and maintenance cost, C scr is the scrap disposal cost, C res is the recovery salvage value, C sub is the subsidy allowance, C rst is the comprehensive benefit of delaying grid upgrade.

[0136] Specifically, the initial cost of the battery energy storage system is related to the number of batteries installed, the capacity, and the power. The initial cost of the battery energy storage system can be calculated as follows:

[0137]

[0138] In the formula, N BESS represents the number of batteries installed, E BESS,m and P BESS,m represent the installed capacity and power of the mth BESS respectively; α and β represent the unit capacity cost and unit power cost of the BESSs.

[0139] The update and replacement cost of the battery energy storage system can be calculated as follows:

[0140]

[0141] In the formula, η represents the installed battery discount rate, T represents the service life of the battery energy storage system, n represents the number of battery cycles, and k represents the lump-sum payment coefficient.

[0142] The charge and discharge cost of the battery energy storage system can be calculated as follows:

[0143]

[0144] In the formula, Rd represents the typical scenario of the distribution network clustered by step 2, Dr represents the number of days occupied by the rth scenario in a year, and h(t) is the difference between the electricity purchase and sale prices during the day based on the time-of-use electricity price.

[0145] The operation and maintenance cost of the battery energy storage system can be calculated as follows:

[0146] C om = C inv × 5% (18)

[0147] That is, the operation and maintenance cost of the battery energy storage system takes 5% of the system cost.

[0148] The scrap disposal cost of the battery energy storage system can be calculated as follows:

[0149] C scr = (α·E pscr + β·E escr ) (19)

[0150] In the formula, E pscr and E escr respectively represent the scrap disposal costs per unit power and per unit capacity.

[0151] The carbon emission reduction benefit brought by installing the battery energy storage system can be calculated as follows:

[0152]

[0153] In the formula, e c is the carbon tax unit price of the node.

[0154] In summary, the comprehensive cost of the battery energy storage system can be expressed as:

[0155] f1 = C Ess = C inv + C rep + C ps + C om + C scr - C res - C sub - C rst (21)

[0156] Furthermore, based on f1 and f2, an objective optimization model for energy storage site selection and capacity determination is established as shown in the following formula. By solving the model with the goal of minimizing f1 and f2, an optimal configuration plan with the lowest comprehensive cost and the best system vulnerability can be determined.

[0157]

[0158] In the formula: f1(x) and f2(x) are the optimization spaces composed of two optimization objective functions, and w(x) and v(x) are the constraint conditions of the functions respectively, constituting the feasible region for function solution. For example, constraints such as the operation safety state constraint of the distribution network, the power flow state constraint, the energy storage system capacity constraint, the safe operation constraint, and other constraint conditions can be considered.

[0159] In this embodiment, by establishing an objective optimization model for energy storage site selection and capacity determination, the impact of energy storage configuration on carbon emission reduction, the comprehensive cost, and the improvement of system resilience by energy storage access can be comprehensively considered.

[0160] Specifically, the following constraint conditions can be configured:

[0161] The power balance constraint is:

[0162]

[0163] where n pv are the installation numbers of photovoltaic respectively; P pv,m (t) are the active power outputs of the m-th photovoltaic at time t; P load (t) is the total load at time t.

[0164] The power flow constraint is:

[0165]

[0166] In the formula, P i (t), Q i (t) are the active and reactive powers injected into the node respectively, and V i (t) is the voltage of node i at time t;

[0167] The voltage constraint is:

[0168]

[0169] In the formula, and are the upper and lower voltage limits of node i respectively.

[0170] The total active and reactive power limit constraints injected by the superior power grid into the distribution network are:

[0171]

[0172] wherein, respectively represent the upper and lower limits of the active and reactive power of the tie line;

[0173] The charge and discharge constraints of the battery system are:

[0174]

[0175] The battery SOC constraint is:

[0176] SOC min ≤SOC i (t)≤SOC max (28)

[0177] The constraints on the installation location and number of batteries are:

[0178] L BESS，i ∈N nodes and L BESS，i ≠L grid

[0179] N BESS，min ≤N BESS，i ≤N max，max (29)

[0181] In a specific application embodiment, as Figure 2 shown, first, by obtaining photovoltaic output data, generator output data, load data, etc., clustering the photovoltaic and load data to obtain typical daily photovoltaic and load curves; then establishing a node carbon emission model according to formulas (3) and (4) and establishing a line comprehensive vulnerability model f2 according to formula (13), and at the same time establishing a full-cycle model f1 of energy storage according to formula (14), establishing an objective optimization model for energy storage site selection and capacity determination based on f1 and f2, and then using an improved particle swarm algorithm to solve the model to solve the optimal installation location and optimal configuration capacity of the energy storage system.

[0182] In a specific application embodiment, the typical distribution network IEEE33-node model can be used. Input the line parameters, load level and network topology connection relationship, system node voltage and branch current limit, and set the initial values of the system base voltage and base power in the IEEE33 nodes. Input the quantization parameters of the energy storage system, including unit price, service life of the battery energy storage system, discount rate, time-of-use electricity price, operation and power constraint conditions into the matlab model, and use the improved particle swarm algorithm to solve the model to obtain the pareto optimal solution set, which is the optimal installation location and optimal configuration capacity of the energy storage system. The topological structure of the simulation analysis using the typical IEEE33-node distribution network system is as Figure 3 shown.

[0183] Specifically, as Figure 4As shown in the figure, the detailed steps of using the improved particle swarm algorithm to solve the target optimization model in this embodiment include:

[0184] Step S301. Initialize the particle population.

[0185] Input the original grid information, such as the type of nodes, the original power output of power sources, load power, node connection status, line reactance data, reactance of photovoltaic power connection lines, etc. After that, initialize the population, including population size, particle dimension, learning factor, upper and lower limits of velocity, upper and lower limits of position, initial position and velocity of particles, and the maximum number of iterations, etc. At the same time, set the energy storage battery capacity as a continuous variable.

[0186] The sequence generated by Tent chaotic mapping has characteristics such as initial sensitivity, ergodicity, and randomness. In a specific application embodiment, the Tent chaotic mapping method can be used to initialize the population to generate a more uniformly distributed population. The definition of chaotic mapping is:

[0187]

[0188] Step S302. Guide the randomly generated particle population using the initial chaotic mapping method and calculate the optimal fitness and global fitness of the population;

[0189] Step S303. Calculate and update the velocity and position information of the particle population according to the status information of each distribution network node in the target area;

[0190] Step S304. Calculate the fitness of the objective function based on the target optimization model and determine whether the constraint conditions are satisfied. If not, return to execute the update of the population velocity and position information and recalculate the optimal fitness of the population; if the constraint conditions are satisfied, select the individual extreme value and the group extreme value through the fitness of the particles. If the fitness of the target particle is better than the global optimal fitness, use the best fitness value of the target particle to update the individual extreme value and the group extreme value of the particle;

[0191] Step S305. Determine whether the maximum preset iteration condition is reached. If so, output the final solution result; otherwise, return to execute the calculation and update of the velocity and position information of the particle population;

[0192] Step S306. Select the optimal solution according to the solution result, and determine the optimal installation position and optimal capacity configuration of the energy storage system.

[0193] Specifically, weights can be established based on information entropy to dynamically update the solution set in step 304, and the solution set is adjusted through the position and velocity update formulas of each particle. Among them, the velocity update is based on the weighted average of the historical optimal position and the global optimal position of the particle. The TOPSIS method is used to select the optimal solution, that is, the solution set with the optimal objective function value in the current particle swarm, to determine the optimal installation location and capacity configuration of the energy storage system.

[0194] In this embodiment, by adopting the above solution method, since the particles are randomly generated, the initial chaotic mapping can make the particles randomly distributed in a more uniform form, making the search more uniform and comprehensive, thereby improving the search accuracy and efficiency.

[0195] In summary, by considering the impact of new energy and energy storage access on node carbon emission reduction, this invention constructs a model for the impact of cluster energy storage on the carbon flow rate of the distribution network to calculate the carbon emission reduction potential of each node. Furthermore, according to the carbon emission potential of each node, the planned locations of energy storage are clustered and partitioned to pre-screen the energy storage installation areas. At the same time, considering the impact of new energy and energy storage access on carbon emission reduction, a comprehensive line vulnerability model is established to identify weak nodes in the distribution network. Combining the comprehensive line vulnerability model with the comprehensive operation cost model, an optimization model for the location and capacity determination of the energy storage system is established, which can achieve the optimal configuration of energy storage considering carbon emission reduction constraints and the improvement of distribution network resilience, obtain the optimal location of energy storage, and at the same time, on the premise of minimizing the economic investment of the energy storage system and minimizing voltage and load fluctuations, optimize the economic model of the energy storage system to obtain the optimal capacity configuration of the energy storage system.

[0196] This embodiment further provides a computer device, including a processor and a memory. The memory is used to store a computer program, and the processor is used to execute the computer program to execute the method as described above.

[0197] It can be understood that the above method of this embodiment can be executed by a single device, such as a computer or a server, etc., or can also be applied to a distributed scenario where multiple devices cooperate with each other to complete. In the case of a distributed scenario, one of the multiple devices can only execute one or more steps of the above method of this embodiment, and the multiple devices interact with each other to complete the above method. The processor can be implemented in the form of a general-purpose CPU, a microprocessor, an application-specific integrated circuit, or one or more integrated circuits, etc., and is used to execute relevant programs to implement the above method of this embodiment. The memory can be implemented in the form of a read-only memory ROM, a random access memory RAM, a static storage device, and a dynamic storage device, etc. The memory can store an operating system and other application programs. When implementing the above method of this embodiment through software or firmware, the relevant program codes are stored in the memory and are called and executed by the processor.

[0198] This embodiment further provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the above method is implemented.

[0199] Those skilled in the art should understand that the above embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes. The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks. These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks. These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0200] The above are only the preferred embodiments of the present invention, and do not impose any form of limitation on the present invention. Although the present invention has been disclosed above with the preferred embodiments, it is not intended to limit the present invention. Therefore, any simple modification, equivalent change, and modification made to the above embodiments according to the technical essence of the present invention without departing from the technical solution of the present invention shall fall within the scope of the technical solution of the present invention.

Claims

1. A method for optimizing the configuration of a distributed energy storage system in a distribution network, characterized in that the steps include: Acquire status information of each distribution network node in the target area, the status information including node connection status, power output data and load data; Calculate the carbon emissions of each distribution network node according to the status information of each distribution network node, and calculate the impact parameters of cluster energy storage on carbon emission reduction of each distribution network node according to the carbon emissions of each distribution network node, and perform cluster partitioning on the planned location of energy storage according to the impact parameters to pre-screen the energy storage system installation area; A calculation model for comprehensive line vulnerability is established by weighting a plurality of fragile features used to characterize the fragile characteristics of energy storage system lines, so as to identify the weak node locations of the target energy storage system, and establish a target optimization model for energy storage site selection and capacity determination with the goal of minimizing the operating cost and optimizing the comprehensive vulnerability of the lines. The state information of each distribution network node in the target area is input to solve the target optimization model, so as to obtain the optimal installation location in the energy storage system installation area and the optimal capacity configuration of the energy storage system.

2. The method for optimizing the configuration of a distributed energy storage system in a distribution network according to claim 1, characterized in that: The carbon emissions of distribution network nodes are calculated according to the following formula: Among them, P l is the active power flowing into node i from the branch; ρ l is the carbon flow density of branch l; G is the set of all branches into which active power flows; i Inject power for the equivalent unit of the i-th node; P bi is the sum of the output power of energy storage and new energy; P N is the node active flux matrix; P B is the branch power flow distribution matrix; P G is the injection distribution matrix of energy storage and new energy units. When the equivalent load P le Less than 0, corresponding to P bi Equal to P le The absolute value of E N is the node carbon potential column vector, indicating the carbon potential of each node, R L represents the carbon flow rate of the node load; EG is the carbon emission intensity of the equivalent unit e gi An N-order column vector.

3. The method for optimizing the configuration of a distributed energy storage system in a distribution network according to claim 1, characterized in that: The impact parameters of cluster energy storage on carbon emission reduction at each distribution network node are calculated according to the following formula: Among them, ΔR ij The parameter representing the impact of the node carbon emissions from node i to node j on the carbon emissions of the power grid, ΔP l The contribution of cluster energy storage to the load power of this node; e l is the N-1 order unit transverse quantity, e li is the N-1th order horizontal quantity with the i-th bit being 1 and the rest being 0; E N is the column vector composed of the carbon potential of each node before being affected by cluster energy storage; Priority to ΔR ij The nodes with a value less than the preset threshold are powered, and the surplus is transmitted to other nodes in the cluster, and the ΔR ij The value is used to partition the planned locations of energy storage into clusters.

4. The method for optimizing the configuration of a distributed energy storage system in a distribution network according to claim 3, characterized in that: According to ΔR ij The steps for cluster partitioning the planned locations of energy storage include: Calculate ΔR between all nodes ij The influence matrix is ​​constructed based on the influence matrix and network characteristics, and a weighted graph model is constructed, where the nodes represent the devices in the power grid and the edge weights are ΔR ij The value is used to indicate the carbon emission impact between nodes; Use community detection algorithm or clustering algorithm to partition each node, so as to gather nodes with similar carbon emission impact into one partition; Determine the installation area of ​​the energy storage system in the preliminary plan based on the zoning results.

5. The method for optimizing the configuration of a distributed energy storage system in a distribution network according to claim 1, characterized in that: The fragile characteristics include system structure fragile characteristics for characterizing system structure fragility and system state fragile characteristics for characterizing system state fragility. The system structure fragile characteristics are constructed using the electrical intermediates of the lines in the system. The system state fragile characteristics include a system power level coefficient for characterizing the system power level state, a system impact transfer ratio for characterizing the impact of line breaks in the system on the power distribution of other lines in the network, and a system voltage deviation rate for characterizing the deviation of the voltage level of each line in the system from the reference rated voltage level.

6. The method for optimizing the configuration of a distributed energy storage system in a distribution network according to claim 5, characterized in that: The calculation expression of the system structure vulnerability characteristic is: Among them, f B is the vulnerability characteristic of the system line structure, B ave is the average value of the electrical intermediate value of the line in the system after the distributed generation is connected to the system, B ave.max For B ave The maximum value of; m is the number of lines; l is the line set; B i is the electrical intermediate number of line i; B e (m,n) is the electrical intermediate value of line (m,n), m and n are the two endpoints of the line, G and D are the power source and load node sets, respectively; P g and P d are the active powers of the power source node g and the load node d respectively; Y gd is the equivalent admittance between node g and node d; N g and N d are the number of power supply and load nodes respectively; I gd (m,n) represents the current value caused on the line (m,n) after a unit current is applied between the power-load node pair (g,d); The calculation expression of the system power level coefficient is: Among them, f L is the system line power level coefficient; L ave is the average value of the line power level coefficient in the system after the distributed generation is connected to the system, L ave.max For L ave The maximum value of L i is the power level coefficient of line i; P i is the active power currently transmitted by line i; P imax is the limit transmission power of line i; m is the number of lines; l is the line set; The calculation expression of the system impact transfer ratio is: Among them, f G is the system shock transfer ratio, G ave is the average value of the line power level coefficient in the system after the distributed generation is connected to the system, G ave.max G ave The maximum value of; m is the number of lines; l is the line set; G i is the impulse transfer ratio coefficient of line i; G i is the impulse transfer ratio of line i; ΔP j is the power change of line j caused by line i disconnection; P j is the original power of line j before line i is disconnected; N L is the total number of lines; The calculation expression of the system voltage deviation rate is: Among them, f DU is the system line voltage deviation rate; △U ave is the average value of the line voltage deviation rate in the system after the distributed generation is connected to the system, △U ave.max △U ave The maximum value of; m is the number of lines; l is the line set; △U i is the voltage deviation rate of line i, ΔU% is the voltage deviation rate of the line; U i , U j are the voltages at the nodes at both ends of a line; U cr is the reference rated voltage; ΔU lim is the maximum allowable voltage offset.

7. The method for optimizing the configuration of a distributed energy storage system in a distribution network according to any one of claims 1 to 6, characterized in that: The calculation model of the comprehensive toughness of the line established is: <h2 style=";text-align:left;direction:ltr">f2 = w1f<h2 style=";text-align:left;direction:ltr"> B <h2 style=";text-align:left;direction:ltr"> +w2f<h2 style=";text-align:left;direction:ltr"> L <h2 style=";text-align:left;direction:ltr"> +w3f<h2 style=";text-align:left;direction:ltr"> G <h2 style=";text-align:left;direction:ltr"> +w4f<h2 style=";text-align:left;direction:ltr"> ΔU Among them, f2 represents the fragile characteristics of the system structure, f L 、f G With f △U They represent the system power level coefficient, system impulse transfer ratio and system voltage deviation rate respectively, and w1~w4 are weight coefficients; The calculation model of operating cost is: f1=C Ess =C inv +C rep +C ps +C om +C scr -C res -C sub -C rst Among them, C Ess is the total cost, C inv is the initial cost, C rep is the replacement cost, C ps is the charge and discharge cost, C om is the system operation and maintenance cost, C scr is the scrapping cost, C res To recover the residual value, C sub It is a subsidy, C rst It is to delay the comprehensive benefits of grid upgrades; The target optimization model established is: min[f1,f2].

8. The method for optimizing the configuration of a distributed energy storage system in a distribution network according to claim 7, characterized in that: It also includes setting any one or more of power balance constraints, power flow constraints, voltage constraints, battery system charge and discharge constraints, battery SOC constraints, and battery installation position and number constraints, wherein the power balance constraints are: Among them, n pv are the number of photovoltaic installations; P pv,m (t) are the active power output of the mth photovoltaic at time t; P load (t) is the total load at time t; The power flow constraint is: Where P i (t), Q i (t) are the active and reactive power injected into the node, V i (t) is the voltage of node i at time t; The voltage constraint is: In i min ≤V i ≤V i max In the formula, and are the upper and lower limits of the voltage at node i respectively; The total active and reactive power limit constraints injected into the distribution network by the upper power grid are: in, Respectively represent the upper and lower limits of the active and reactive power of the tie line; The battery system charge and discharge constraints are: The battery SOC constraint is: SOC min ≤SOC i (t)≤SOC max The battery installation position and number constraints are: L BESS，i ∈N nodes And L BESS，i ≠L grid N BESS,min ≤N BESS，i ≤N max,max 。 9. The method for optimizing the configuration of a distributed energy storage system in a distribution network according to any one of claims 1 to 6, characterized in that: The target optimization model is solved by using an improved particle swarm algorithm, and the steps include: Initialize particle population; Randomly generate particle populations and calculate the optimal fitness and global fitness of the population; According to the status information of each distribution network node in the target area, the speed and position information of the particle population is calculated and updated; Based on the target optimization model, the fitness of the objective function is calculated and whether the constraint conditions are met is determined. If not, the population speed and position information are updated and the optimal fitness of the population is recalculated. If the constraint conditions are met, the individual extreme value and the group extreme value are selected according to the fitness of the particle. If the fitness of the target particle is better than the global optimal fitness, the individual extreme value and the group extreme value of the particle are updated using the best fitness value of the target particle. Determine whether the maximum preset iteration condition is reached, if so, output the final solution result, otherwise return to execute the calculation and update the speed and position information of the particle population; The optimal solution is selected based on the solution results to determine the optimal installation location and optimal capacity configuration of the energy storage system.

10. A system for optimizing the configuration of a distributed energy storage system in a distribution network, comprising a processor and a memory, wherein the memory is used to store a computer program, characterized in that: The processor is configured to execute the computer program to perform the method according to any one of claims 1 to 9.

Images