A method for evaluating adjustable capacity of a power distribution network and a terminal

Abstract

The application discloses a kind of distribution network adjustable capacity evaluation method and terminal, using clustering algorithm based on wind power related data, photovoltaic related data, energy storage related data and electric vehicle related data respectively to cluster distributed power supply, distributed energy storage and electric vehicle are divided into cluster, obtain distributed power supply cluster, distributed energy storage cluster and electric vehicle cluster, establish distributed power supply cluster adjustable power calculation model, distributed energy storage cluster adjustable power calculation model and electric vehicle cluster adjustable power calculation model to distributed power supply cluster, distributed energy storage cluster and electric vehicle cluster, aggregation is carried out to three kinds of adjustable power calculation model, obtain the adjustable range of regional distribution network flexibility resource and adjustable upper and lower limit constraint, to accurately evaluate the adjustable capacity of distribution network, it is helpful to provide basis and guidance for flexibility resource to participate in peak shaving and frequency modulation, effectively reduce peak-valley difference, and suppress load fluctuation, maintain grid smooth operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power distribution network regulation, in particular to a method for evaluating the adjustable capacity of a power distribution network and a terminal. BACKGROUND

[0002] The penetration rate of new energy in the power system is increasing, greatly increasing the uncertainty factors in the power grid. These resources are affected by weather conditions and time changes, have large output fluctuations, and are difficult to accurately predict and control, leading to unstable power grids, which require flexible resources to balance the power system. Load side also often fluctuates, and the peak-valley difference can cause the power grid to overload or operate inefficiently. Flexible resources can be used to adjust power supply to meet different time load demands, thereby improving the reliability and efficiency of the power grid.

[0003] Although there are various potential flexible resources, traditional power system evaluation methods often cannot fully quantitatively evaluate their participation and effectiveness, i.e., existing methods cannot effectively consider various flexible resources such as distributed power sources, distributed energy storage, and electric vehicles to accurately evaluate the adjustable capacity of the power distribution network. Therefore, there is an urgent need for a new method to quantitatively evaluate the potential of these resources to better manage the dispatchability and stability of the power system. SUMMARY

[0004] The technical problem to be solved by the present application is to provide a method for evaluating the adjustable capacity of a power distribution network, which can accurately evaluate the adjustable capacity of the power distribution network.

[0005] To solve the above technical problems, the present application adopts a technical solution:

[0006] A method for evaluating the adjustable capacity of a power distribution network, comprising the steps of:

[0007] Obtaining wind power related data, photovoltaic related data, energy storage related data, and electric vehicle related data in the power distribution network;

[0008] Using a clustering algorithm to cluster and divide distributed power sources, distributed energy storage, and electric vehicles based on the wind power related data, the photovoltaic related data, the energy storage related data, and the electric vehicle related data, respectively, to obtain distributed power source clusters, distributed energy storage clusters, and electric vehicle clusters;

[0009] Based on the wind power related data, the photovoltaic related data, the energy storage related data, and the electric vehicle related data, respectively, a distributed power source cluster adjustable power calculation model, a distributed energy storage cluster adjustable power calculation model, and an electric vehicle cluster adjustable power calculation model are established for the distributed power source clusters, the distributed energy storage clusters, and the electric vehicle clusters;

[0010] The distributed power source cluster adjustable power calculation model, the distributed energy storage cluster adjustable power calculation model and the electric vehicle cluster adjustable power calculation model are aggregated to obtain an adjustable range and adjustable upper and lower limit constraints of a flexible resource of a regional power distribution network.

[0011] To solve the above technical problems, another technical solution adopted by the present application is:

[0012] A power distribution network adjustable capacity evaluation terminal comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the following steps when executing the computer program:

[0013] Obtain wind power related data, photovoltaic related data, energy storage related data and electric vehicle related data in the power distribution network;

[0014] Cluster the distributed power source, distributed energy storage and electric vehicle based on the wind power related data, photovoltaic related data, energy storage related data and electric vehicle related data respectively using a clustering algorithm to obtain a distributed power source cluster, a distributed energy storage cluster and an electric vehicle cluster;

[0015] Establish a distributed power source cluster adjustable power calculation model, a distributed energy storage cluster adjustable power calculation model and an electric vehicle cluster adjustable power calculation model based on the wind power related data, photovoltaic related data, energy storage related data and electric vehicle related data for the distributed power source cluster, the distributed energy storage cluster and the electric vehicle cluster respectively;

[0016] The distributed power source cluster adjustable power calculation model, the distributed energy storage cluster adjustable power calculation model and the electric vehicle cluster adjustable power calculation model are aggregated to obtain an adjustable range and adjustable upper and lower limit constraints of a flexible resource of a regional power distribution network.

[0017] The beneficial effects of the present application are that: the clustering algorithm is used to perform cluster division on the distributed power supply, distributed energy storage and electric vehicles based on wind power related data, photovoltaic related data, energy storage related data and electric vehicle related data, to obtain a distributed power supply cluster, a distributed energy storage cluster and an electric vehicle cluster, and the distributed power supply cluster, the distributed energy storage cluster and the electric vehicle cluster are established based on the wind power related data, the photovoltaic related data, the energy storage related data and the electric vehicle related data to establish a distributed power supply cluster adjustable power calculation model, a distributed energy storage cluster adjustable power calculation model and an electric vehicle cluster adjustable power calculation model, and the distributed power supply cluster adjustable power calculation model, the distributed energy storage cluster adjustable power calculation model and the electric vehicle cluster adjustable power calculation model are aggregated to obtain the adjustable range and the adjustable upper and lower limit constraints of the regional distribution network flexibility resource, the present application considers new energy power generation, distributed energy storage, electric vehicles and other adjustable resources, and uses the clustering algorithm combined with the adjustable power calculation model to evaluate the adjustable capacity of the regional distribution network, so as to accurately evaluate the adjustable capacity of the distribution network, which helps to provide basis and guidance for the flexible resources participating in peak shaving and frequency modulation, effectively reduces the peak-valley difference, and stabilizes the load fluctuation and maintains the stable operation of the power grid. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 A step flow chart of a distribution network adjustable capacity evaluation method according to an embodiment of the present application;

[0019] Figure 2 A structural schematic diagram of a distribution network adjustable capacity evaluation terminal according to an embodiment of the present application. DETAILED DESCRIPTION

[0020] To explain the technical content, the purposes and effects of the present application in detail, the following will be described in conjunction with the embodiments and the accompanying drawings.

[0021] Please refer to Figure 1 A distribution network adjustable capacity evaluation method, comprising the steps of:

[0022] Obtaining wind power related data, photovoltaic related data, energy storage related data and electric vehicle related data in a distribution network;

[0023] Using a clustering algorithm to perform cluster division on distributed power supply, distributed energy storage and electric vehicles based on the wind power related data, the photovoltaic related data, the energy storage related data and the electric vehicle related data, to obtain a distributed power supply cluster, a distributed energy storage cluster and an electric vehicle cluster;

[0024] establish a distributed power source cluster adjustable power calculation model, a distributed energy storage cluster adjustable power calculation model and an electric vehicle cluster adjustable power calculation model based on the wind power related data, the photovoltaic related data, the energy storage related data and the electric vehicle related data respectively;

[0025] The distributed power source cluster adjustable power calculation model, the distributed energy storage cluster adjustable power calculation model and the electric vehicle cluster adjustable power calculation model are aggregated to obtain the adjustable range and the adjustable upper and lower limit constraints of the flexible resource of the regional power distribution network.

[0026] From the above description, the beneficial effects of the present application are that: the clustering algorithm is used to divide the distributed power source, the distributed energy storage and the electric vehicle based on the wind power related data, the photovoltaic related data, the energy storage related data and the electric vehicle related data respectively to obtain the distributed power source cluster, the distributed energy storage cluster and the electric vehicle cluster, the distributed power source cluster adjustable power calculation model, the distributed energy storage cluster adjustable power calculation model and the electric vehicle cluster adjustable power calculation model are established based on the wind power related data, the photovoltaic related data, the energy storage related data and the electric vehicle related data respectively, and the distributed power source cluster adjustable power calculation model, the distributed energy storage cluster adjustable power calculation model and the electric vehicle cluster adjustable power calculation model are aggregated to obtain the adjustable range and the adjustable upper and lower limit constraints of the flexible resource of the regional power distribution network, the present application considers the new energy generation, the distributed energy storage and the electric vehicle, and the adjustable ability of the regional power distribution network is evaluated by using the clustering algorithm combined with the adjustable power calculation model, so that the adjustable ability of the power distribution network is accurately evaluated, which helps to provide basis and guidance for the flexible resource participating in peak shaving and frequency modulation, effectively reduces the peak-valley difference, suppresses the load fluctuation and maintains the stable operation of the power grid.

[0027] Further, the distributed power source cluster, the distributed energy storage cluster and the electric vehicle cluster obtained by using the clustering algorithm based on the wind power related data, the photovoltaic related data, the energy storage related data and the electric vehicle related data respectively include:

[0028] The Fast Unfolding clustering algorithm is used to divide the distributed power source based on the wind power related data and the photovoltaic related data to obtain the distributed power source cluster;

[0029] The fuzzy C-means clustering algorithm is used to divide the distributed energy storage based on the energy storage related data to obtain the distributed energy storage cluster;

[0030] The improved k-means clustering algorithm is used to cluster the electric vehicles based on the electric vehicle related data, so as to obtain electric vehicle clusters.

[0031] As can be seen from the above description, different flexible resources are clustered by different clustering algorithms, which is more suitable for the characteristics of flexible resources, and reliable and accurate clustering is realized, so as to realize accurate adjustable capacity evaluation subsequently.

[0032] Further, the Fast Unfolding clustering algorithm is used to cluster the distributed power based on the wind power related data and the photovoltaic related data, so as to obtain distributed power clusters, including:

[0033] The net load of the distributed power node of the power distribution network is calculated according to the distributed power output of the distributed power node and the load, and the net load of the distributed power node is taken as a first clustering index;

[0034] The reactive power regulation capacity and the active power regulation capacity of the distributed power are taken as a second clustering index;

[0035] The first clustering index and the second clustering index are normalized respectively, so as to obtain a normalized first clustering index and a normalized second clustering index;

[0036] The similarity between the distributed power nodes is calculated based on the normalized first clustering index and the normalized second clustering index;

[0037] Any cluster in which a distributed power node is located is randomly selected and merged with other clusters to obtain a new cluster;

[0038] The increment value of a modularity function corresponding to the new cluster is calculated based on the similarity between the distributed power nodes;

[0039] If the maximum increment value of the current modularity function is greater than a first preset value, the cluster in which the distributed power node is located is merged with the cluster corresponding to the maximum increment value of the modularity function to obtain a new cluster;

[0040] If the maximum increment value of the current modularity function is not greater than the first preset value, the cluster in which the distributed power node is located is not merged with the cluster corresponding to the maximum increment value of the modularity function;

[0041] The cluster is taken as a new distributed power node, and the step of calculating the similarity between the distributed power nodes based on the normalized first clustering index and the normalized second clustering index is executed, until the maximum increment value of the modularity function no longer changes, so as to obtain distributed power clusters.

[0042] From the above description, the FastUnfolding clustering algorithm is used to divide the distributed power clusters. When the clustering algorithm divides the clusters, the number of clusters does not need to be given in advance. The optimal number of clusters is obtained by optimizing the modularity function value. The modularity function value of the clustering algorithm is small in different running scenarios, which has strong adaptability and ensures the effect of the cluster division of the distributed power.

[0043] Further, the improved k-means clustering algorithm is used to divide the electric vehicles into clusters based on the electric vehicle related data to obtain the electric vehicle clusters.

[0044] K electric vehicle nodes are randomly selected from the power distribution network as initial cluster centers;

[0045] The distances between the remaining electric vehicle nodes and the K cluster centers are calculated;

[0046] An evaluation index of the clustering effect is determined according to the universal gravitation model;

[0047] The remaining electric vehicle nodes corresponding to the maximum value of the evaluation index are attributed to the cluster centers, and the cluster centers are updated;

[0048] It is judged whether the evaluation index reaches a second preset value or the current iteration number reaches a preset number. If yes, the iteration is stopped, and the electric vehicle clusters are obtained. If not, the step of calculating the distances between the remaining electric vehicle nodes and the K cluster centers is returned.

[0049] From the above description, when the electric vehicles are divided into clusters, the k-means clustering algorithm has simple principles and is easy to implement. The data can be clustered according to multiple features or dimensions. The universal gravitation model can simulate the mutual attraction and repulsion between data points, which helps the algorithm to remain stable when facing noise and outliers, thereby improving the robustness of clustering. The traditional k-means algorithm assumes that the clusters are convex and similar in size. The universal gravitation model can simulate more complex interactions, which can discover clusters of arbitrary shape and size. The universal gravitation model adjusts the cluster centers by simulating the interaction between data points, which can reduce the dependence on the selection of initial cluster centers and reduce the risk of falling into a local optimal solution. Therefore, the universal gravitation model is used to improve the k-means clustering algorithm, and the improved k-means clustering algorithm is used to divide the clusters to obtain the electric vehicle clusters, thereby improving the reliability and accuracy of the electric vehicle cluster division.

[0050] Further, the establishing, based on the wind power related data, the photovoltaic related data, the energy storage related data and the electric vehicle related data, the distributed power source cluster, the distributed energy storage cluster and the electric vehicle cluster, a distributed power source cluster adjustable power calculation model, a distributed energy storage cluster adjustable power calculation model and an electric vehicle cluster adjustable power calculation model respectively comprises:

[0051] establishing, based on the wind power related data, a wind power cluster adjustable power calculation model based on the distributed power source cluster;

[0052] establishing, based on the photovoltaic related data, a photovoltaic cluster adjustable power calculation model based on the distributed power source cluster;

[0053] establishing, based on the energy storage related data, a distributed energy storage cluster adjustable power calculation model based on the distributed energy storage cluster;

[0054] establishing, based on the electric vehicle related data, an electric vehicle cluster adjustable power calculation model based on the electric vehicle cluster.

[0055] As can be seen from the above description, the wind power cluster adjustable power calculation model, the photovoltaic cluster adjustable power calculation model, the distributed energy storage cluster adjustable power calculation model and the electric vehicle cluster adjustable power calculation model are established, and different adjustable power calculation models can comprehensively and accurately reflect the adjustable capacity of the distribution network.

[0056] Further, the establishing, based on the wind power related data, the photovoltaic related data, the energy storage related data and the electric vehicle related data, the distributed power source cluster, the distributed energy storage cluster and the electric vehicle cluster, a distributed power source cluster adjustable power calculation model, a distributed energy storage cluster adjustable power calculation model and an electric vehicle cluster adjustable power calculation model respectively comprises:

[0057] determining, based on the wind power related data, the output power of the wind turbine;

[0058] determining, based on the output power of the wind turbine, the single wind power adjustable potential;

[0059] determining, based on the single wind power adjustable potential, the wind power cluster adjustable power calculation model;

[0060] The establishing, based on the photovoltaic related data, the photovoltaic cluster adjustable power calculation model based on the distributed power source cluster comprises:

[0061] determining, based on the photovoltaic related data, the output power of the photovoltaic generator;

[0062] determining, based on the output power of the photovoltaic generator, the single photovoltaic generator adjustable potential;

[0063] determining, based on the single photovoltaic generator adjustable potential, the photovoltaic cluster adjustable power calculation model.

[0064] According to the above description, the wind power cluster adjustable power calculation model is determined according to the single wind power adjustable potential, the photovoltaic cluster adjustable power calculation model is determined according to the single photovoltaic generator adjustable potential, and the accurate evaluation of the wind power cluster and the photovoltaic cluster adjustable power is realized.

[0065] Further, the distributed energy storage cluster adjustable power calculation model is established based on the distributed energy storage related data, and the establishment of the distributed energy storage cluster adjustable power calculation model based on the distributed energy storage related data includes:

[0066] The energy storage charging and discharging constraints are established;

[0067] The energy storage load group power is determined according to the energy storage related data;

[0068] The distributed energy storage cluster adjustable power calculation model is obtained based on the energy storage load group power and the energy storage charging and discharging constraints.

[0069] According to the above description, the distributed energy storage cluster adjustable power calculation model is obtained based on the energy storage load group power and the energy storage charging and discharging constraints, which can reasonably and effectively calculate the adjustable power of the distributed energy storage.

[0070] Further, the electric vehicle cluster adjustable power calculation model is established based on the electric vehicle related data, and the establishment of the electric vehicle cluster adjustable power calculation model based on the electric vehicle related data includes:

[0071] The battery safety power constraint and the charging and discharging power constraint of the electric vehicle are established;

[0072] The total response potential is determined according to the charging and discharging response potential of the user of the electric vehicle;

[0073] The electric vehicle cluster adjustable power calculation model is obtained according to the total response potential, the battery safety power constraint and the charging and discharging power constraint.

[0074] According to the above description, the electric vehicle cluster adjustable power calculation model is obtained according to the total response potential, the battery safety power constraint and the charging and discharging power constraint, which can calculate the adjustable power of the electric vehicle cluster while ensuring the service life of the battery of the electric vehicle, and ensure the reliable operation of the power distribution network.

[0075] Further, the distributed power source cluster adjustable power calculation model, the distributed energy storage cluster adjustable power calculation model and the electric vehicle cluster adjustable power calculation model are aggregated to obtain the adjustable range and the adjustable upper and lower limit constraints of the regional power distribution network flexibility resource, and the aggregation of the distributed power source cluster adjustable power calculation model, the distributed energy storage cluster adjustable power calculation model and the electric vehicle cluster adjustable power calculation model includes:

[0076] The distributed power source cluster adjustable power calculation model, the distributed energy storage cluster adjustable power calculation model and the electric vehicle cluster adjustable power calculation model are aggregated using the Minkowski method to obtain the adjustable range and adjustable upper and lower limit constraints of the regional power distribution network flexibility resource.

[0077] As can be seen from the above description, the distributed power source cluster adjustable power calculation model, the distributed energy storage cluster adjustable power calculation model and the electric vehicle cluster adjustable power calculation model are aggregated using the Minkowski method, which has high precision and improves the accuracy of the adjustable capacity evaluation of the power distribution network. The adjustable range and adjustable upper and lower limit constraints of the regional power distribution network flexibility resource effectively reflect the adjustable capacity of the power distribution network, so that the power distribution network can more efficiently participate in the interactive operation of the power market, and the unified management and dispatch of the power grid are facilitated.

[0078] Please refer to Figure 2 Another embodiment of the present application provides a power distribution network adjustable capacity evaluation terminal, which comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements each step of the above-mentioned power distribution network adjustable capacity evaluation method when executing the computer program.

[0079] The above-mentioned power distribution network adjustable capacity evaluation method and terminal of the present application can be applied to power distribution networks that need to evaluate adjustable capacity, and the following will be described through a specific embodiment:

[0080] Please refer to Figure 1 The first embodiment of the present application is:

[0081] A power distribution network adjustable capacity evaluation method comprises the following steps:

[0082] S1, obtaining wind power related data, photovoltaic related data, energy storage related data and electric vehicle related data in the power distribution network.

[0083] In an optional embodiment, the wind power related data includes real-time wind speed, the photovoltaic related data includes light intensity and environmental temperature, the energy storage related data includes the rated capacity of the energy storage system, and the electric vehicle related data includes the SOC state of the user of the electric vehicle.

[0084] S2, using a clustering algorithm to cluster and divide distributed power sources, distributed energy storages and electric vehicles based on the wind power related data, the photovoltaic related data, the energy storage related data and the electric vehicle related data to obtain distributed power source clusters, distributed energy storage clusters and electric vehicle clusters, specifically including S21-S23:

[0085] S21. Using the FastUnfolding clustering algorithm, distributed power sources are clustered based on the wind power-related data and the photovoltaic-related data to obtain distributed power source clusters, specifically including S211-S219:

[0086] The Fast Unfolding clustering algorithm does not require a pre-defined number of clusters when partitioning clusters; the optimal number of clusters is obtained by optimizing the modularity function value. Under different operating scenarios, the modularity function value of cluster partitioning based on this method shows minimal variation, demonstrating strong adaptability.

[0087] S211. Calculate the net load of the distributed power generation nodes based on the distributed power output and load of the distributed power generation nodes in the distribution network, and use the net load of the distributed power generation nodes as the first grouping index.

[0088] The net load of a distributed generation node reflects the relationship between the output of the distributed generation connected to the node and the load. Specifically, the calculation of the net load of the distributed generation node based on the output and load of the distributed generation nodes in the distribution network is as follows:

[0089] P net_i =P N_i -P L_i ;

[0090] In the formula, P net_i P represents the net load of distributed power node i. N_i P represents the distributed power output of distributed power node i. L_i This represents the load of distributed power node i.

[0091] The output range of distributed power sources is:

[0092] P N_i ∈[P N_i,min ,P N_i,max ];

[0093] In the formula, P N_i,min P represents the lower limit of the distributed power output of distributed power node i. N_i,max This represents the upper limit of the distributed power output of distributed power node i.

[0094] S212. The reactive power regulation capacity and active power regulation capacity of distributed power sources are used as the second grouping index.

[0095] The reactive power regulation capacity and active power regulation capacity of the distributed power source are as follows:

[0096]

[0097] In the formula, Qc P represents the reactive power regulation capacity of distributed power sources. c Q represents the active power regulation capacity of distributed generation. min Q represents the minimum reactive power regulation capacity. max P represents the maximum reactive power regulation capacity. max This indicates the maximum active power regulation capacity.

[0098] The second grouping index can be used to group distributed power sources with similar active and reactive power regulation capacities into the same cluster.

[0099] S213. Normalize the first cluster index and the second cluster index respectively to obtain the normalized first cluster index and the normalized second cluster index, specifically:

[0100]

[0101] In the formula, x i ′ ,m Let x represent the m-th cluster index of the i-th node after normalization. i,m Let x represent the m-th cluster index of the i-th node. i,m,min Let x represent the minimum value among the M clustering indices of the i-th node. i,m,max This represents the maximum value among the M clustering indicators of the i-th node.

[0102] S214. Calculate the similarity between distributed power nodes based on the normalized first cluster index and the normalized second cluster index, so that distributed power nodes within the same cluster have similar operating characteristics, specifically:

[0103]

[0104] In the formula, s ij Let R represent the similarity between distributed power node i and distributed power node j, R represent the weighted maximum value of the product of the indices of any two nodes, M represent the number of cluster indices, and α represent the similarity between them. m Let x′ represent the weight of the m-th cluster indicator. j,m This represents the m-th cluster index of the j-th node after normalization.

[0105] S215. Randomly select any distributed power node's cluster and merge it with other clusters to obtain a new cluster.

[0106] S216. Calculate the incremental value of the modularity function corresponding to the new cluster based on the similarity between the distributed power nodes.

[0107] The modularity function measures the rationality of cluster partitioning, specifically as follows:

[0108]

[0109] In the formula, Q represents modularity, s represents the sum of similarities between two nodes, s i represents the sum of similarities between distributed power supply node i and other distributed power supply nodes, s j represents the sum of similarities between distributed power supply node j and other distributed power supply nodes, represents the relationship between distributed power supply node i and distributed power supply node j, 1 if in the same cluster, and 0 if not in the same cluster.

[0110] The incremental value AQ of the modularity function corresponding to the new cluster is the incremental value of the modularity function corresponding to the cluster of the current iteration and the modularity function corresponding to the cluster of the last iteration. The incremental value of the modularity function of each iteration is recorded, and then a maximum incremental value appears in each iteration.

[0111] S217, if the maximum incremental value AQ of the current modularity function is greater than the first preset value, the cluster in which any distributed power supply node is located is merged with the cluster corresponding to the maximum incremental value of the modularity function to obtain a new cluster. max

[0112] In an optional embodiment, the first preset value is 0.

[0113] S218, if the maximum incremental value of the current modularity function is not greater than the first preset value, the cluster in which any distributed power supply node is located is not merged with the cluster corresponding to the maximum incremental value of the modularity function.

[0114] S219, taking the cluster as a new distributed power supply node, returning to execute S214 until the maximum incremental value of the modularity function no longer changes, and obtaining a distributed power supply cluster.

[0115] Taking the cluster as a new distributed power supply node, returning to execute S214, and then S214 calculates the similarity between two clusters.

[0116] S22, using a fuzzy C-means clustering algorithm to perform cluster division on distributed energy storage based on the energy storage related data, and obtaining a distributed energy storage cluster, specifically including S221-S225:

[0117] The k-means clustering algorithm only hard assigns data points to the nearest cluster center, while the fuzzy C-means clustering algorithm can assign a fuzzy degree value to each data point, which can better cope with noise and fuzziness in the data set, and the fuzzy C-means clustering algorithm is relatively insensitive to the selection of initial cluster centers, and therefore can obtain a better clustering result. ​

[0118] The objective function of the fuzzy C-means clustering algorithm is specifically:

[0119]

[0120] J = Σ m (U, P) represents a weighted sum of squares, m represents a weighted exponent, U represents a membership matrix, P represents a clustering center, N represents a number of nodes accessed in a power system, C represents a number of clusters, μ i,k represents a membership function, i.e., a membership relationship of a sample to a subset, and μ i,k ∈ [0, 1], the membership matrix U = [μ i,k ] c×k , represents a distance of a clustering center to an i-th sample.

[0121] S221, initialize a membership matrix U, a scheduling period T, and a number of clusters C;

[0122] S222, take L iterations, and calculate a clustering center, specifically:

[0123]

[0124] wherein, represents a clustering center of the L+1 iteration, represents a membership function of the L iteration, x k represents a k-th sample.

[0125] S223, correct the membership matrix, specifically:

[0126]

[0127] wherein, represents a distance of a clustering center to a j-th sample.

[0128] S224, calculate an objective function, specifically:

[0129]

[0130] S225, if the above calculation satisfies |μ i,k (L+1) - μ i,k (L) |≤ε, μ i,k =max{μ i,k}, wherein ε represents an error threshold, a very small constant, then the sample is classified into a subset, and a distributed energy storage cluster is obtained, and if the condition is not satisfied, L=L+1, and the process returns to S222.

[0131] S23. Using an improved k-means clustering algorithm, the electric vehicles are clustered based on the relevant data, resulting in electric vehicle clusters, specifically including S231-S235:

[0132] Because the k-means clustering algorithm is simple in principle and easy to implement, and can cluster data based on multiple features or dimensions, this invention uses the k-means clustering algorithm to divide the EV (electric vehicle) population based on its grid entry time, grid exit time, and initial remaining battery power upon entering the grid. The k-means algorithm uses k as a parameter to divide n samples into k clusters, aiming to ensure high similarity among samples within a cluster and low similarity among samples between clusters. The k-means algorithm uses distance as a similarity evaluation metric, and the goal of clustering is to minimize the distance from each cluster center to the observations within that cluster.

[0133] Gravity, a universally present interaction property in nature, is introduced into the k-means clustering model. The gravitational model can simulate the attraction and repulsion between data points, which helps the algorithm remain stable in the face of noise and outliers, thus improving the robustness of clustering. Traditional k-means algorithms assume clusters are convex and of similar scale, while the gravitational model can simulate more complex interactions, potentially discovering clusters of arbitrary shapes and sizes. The gravitational model adjusts cluster centers by simulating the interactions between data points, reducing dependence on the initial cluster center selection and thus lowering the risk of getting trapped in local optima. Therefore, the gravitational model is used to improve the k-means clustering algorithm.

[0134] S231. Randomly select K electric vehicle nodes from the distribution network as the initial cluster centers.

[0135] S232. Calculate the distances between the remaining electric vehicle nodes and the K cluster centers, specifically:

[0136] r i =||x i -c j ||;

[0137] In the formula, r i x represents the distance between electric vehicle node i and the K cluster centers. i Represents electric vehicle nodes i and c j Let j represent the cluster center.

[0138] S233. Determine the evaluation index of clustering effect based on the universal gravitation model, specifically:

[0139]

[0140] EI i is an evaluation index representing clustering effect, V is an evaluation adjustment coefficient, P is load capacity of two points, p i is load capacity of electric vehicle node i, r i is distance from electric vehicle node i to cluster center.

[0141] S234, attribute the remaining electric vehicle node corresponding to the maximum value of the evaluation index to the cluster center, and update the cluster center.

[0142] Specifically, MAX(EI i ) is used to attribute x i to the cluster of c j , and update the cluster center, specifically:

[0143]

[0144] S235, judge whether the evaluation index reaches a second preset value or the current iteration number reaches a preset number, if yes, stop iteration to obtain the electric vehicle cluster, if not, return to execute S232.

[0145] S3, based on the wind power related data, the photovoltaic related data, the energy storage related data and the electric vehicle related data, respectively, establish a distributed power source cluster adjustable power calculation model, a distributed energy storage cluster adjustable power calculation model and an electric vehicle cluster adjustable power calculation model for the distributed power source cluster, the distributed energy storage cluster and the electric vehicle cluster, specifically including S31-S34:

[0146] S31, according to the wind power related data, establish a wind power cluster adjustable power calculation model based on the distributed power source cluster, specifically including S311-S313:

[0147] S311, determine output power of wind turbine according to the wind power related data, active power of wind turbine is related to natural factors such as wind speed and wind direction and control strategy, wherein the most influential factor is real-time wind speed, in order to ensure safe operation of wind turbine, wind speed borne by blade cannot be increased indefinitely, and operation state needs to be cut out if necessary, output power of wind turbine changes according to change of wind speed, specifically:

[0148]

[0149] P v,i is output power of wind turbine, v is real-time wind speed, v in is cut-in wind speed, P N is rated power of wind turbine, v N is rated wind speed, vout cut-out wind speed, when v in , the wind turbine has no power output; when v in ≤ v ≤ v N , the output power of the wind turbine increases with the increase of the wind speed; when v N ≤ v < v out , the wind turbine still outputs rated power, if the wind speed continues to increase beyond the cut-out wind speed, the wind turbine will be disconnected from the grid to avoid accidents.

[0150] To realize the reasonable scheduling of the wind farm output, the wind power output prediction system of the scheduling department should ensure a certain accuracy, that is, the wind farm output prediction error is small enough, specifically:

[0151]

[0152] In the formula, P wp represents the predicted output of the wind farm, P represents the actual output of the wind farm, P wcap represents the capacity of the wind farm, and δ represents the maximum error of the wind farm output prediction.

[0153] S312, determining the individual wind power adjustable potential according to the output power of the wind turbine, specifically:

[0154]

[0155] P v,min,i ≤ ΔP v,i ≤ P v,max,i ;

[0156] In the formula, ΔP v,i represents the individual wind power adjustable potential, P v,i,max (t+1) represents the predicted maximum power of the wind turbine i in the next period, P v,i (t) represents the output power of the wind turbine i at the current period wind speed, P N , i represents the rated power of the wind turbine i, P v,min,i represents the lower limit of the adjustable power of the wind turbine i, and P v,max,i represents the upper limit of the adjustable power of the wind turbine i.

[0157] S313, determining the wind power cluster adjustable power calculation model according to the individual wind power adjustable potential, specifically:

[0158]

[0159] In the formula, ΔP W represents the wind power cluster adjustable power, and n represents the number of wind turbines in the wind power cluster.

[0160] S32, establishing a photovoltaic cluster adjustable power calculation model based on the distributed power cluster according to the photovoltaic related data, specifically comprising S321-S323:

[0161] S321, determining the output power of the photovoltaic unit according to the photovoltaic related data, the output of the solar photovoltaic generator set is mainly affected by the light intensity, the battery junction temperature and the environmental temperature, specifically:

[0162]

[0163] P p,i (t) represents the output power of the photovoltaic unit at time t, P SET represents the rated output power of the photovoltaic unit, S(t) represents the light intensity of the photovoltaic unit at time t, S SET represents the standard light intensity of the photovoltaic unit, k T represents the temperature coefficient, T(t) represents the environmental temperature at time t, T SET represents the standard environmental temperature.

[0164] S322, determining the adjustable potential of a single photovoltaic generator according to the output power of the photovoltaic unit, specifically:

[0165]

[0166] P p,min,i ≤ΔP p,i ≤P p,max,i ;

[0167] P p,i represents the adjustable potential of a single photovoltaic generator, P p,i,max (t+1) represents the predicted maximum power of the photovoltaic unit i in the next period, P p,i (t) represents the output power of the photovoltaic unit i under the current period condition, P N,p,i represents the rated power of the photovoltaic unit i, P p,min,i represents the lower limit of the adjustable power of the photovoltaic unit i, P p,max,i represents the upper limit of the adjustable power of the photovoltaic unit i.

[0168] S323, determining the photovoltaic cluster adjustable power calculation model according to the adjustable potential of a single photovoltaic generator, specifically:

[0169]

[0170] P PV represents the photovoltaic cluster adjustable power, and n represents the number of photovoltaic units.

[0171] S33, establishing a distributed energy storage cluster adjustable power calculation model based on the distributed energy storage cluster according to the energy storage related data, specifically comprising S331-S333:

[0172] S331, establishing an energy storage charging and discharging constraint, specifically:

[0173] The charging and discharging power P of the energy storage system is ES defined as charging positive and discharging negative, assuming that the charging efficiency η c and the discharging efficiency η d of the energy storage system remain unchanged during operation, let:

[0174]

[0175] In the formula, α ES (t) represents the charging and discharging efficiency of the energy storage system at time t, k(t) represents the operating state of the energy storage system, if it is in the charging state, it is 1, if it is in the discharging state, it is -1, if it is in the non-charging and non-discharging state, it is 0, represents the discharging efficiency of the energy storage system;

[0176] The energy storage SOC changes to:

[0177]

[0178] 0≤P ES ≤P ES,c,max ;

[0179] 0≤P ES ≤P ES,d,max ;

[0180] In the formula, SOC ES represents the state of charge of the energy storage system, P ES represents the charging and discharging power of the energy storage system, C ES represents the rated capacity of the energy storage system, P ES,c,max represents the maximum charging power of the energy storage system, P ES,d,max represents the maximum discharging power of the energy storage system;

[0181] The energy storage charging and discharging constraint is established as:

[0182] P ESmin,i ≤SOC ES (t)≤P ESmax,i ;

[0183] In the formula, P ESmin,i represents the minimum state of charge of the energy storage system, P ESmax,i represents the maximum state of charge of the energy storage system, SOC ES(t) represents the state of charge of the energy storage system at time t.

[0184] S332, determining the energy storage load group power according to the energy storage related data, specifically:

[0185]

[0186] wherein, represents the energy storage load group power at time t, N represents the number of energy storages, k j (t) represents the operating state of the energy storage system j, represents the charge and discharge rated power of the energy storage system j.

[0187] S333, obtaining a distributed energy storage cluster adjustable power calculation model based on the energy storage load group power and the energy storage charge and discharge constraints, including:

[0188]

[0189] wherein, ΔP ES (t) represents the distributed energy storage cluster adjustable power at time t, represents the energy storage load group power before adjustment, represents the energy storage load group power after adjustment.

[0190] S34, establishing an electric vehicle cluster adjustable power calculation model based on the electric vehicle cluster according to the electric vehicle related data, specifically including S341-S343:

[0191] S341, establishing the battery safety power constraint and the charge and discharge power constraint of the electric vehicle.

[0192] To ensure the service life of the EV battery, the safety power limit of the battery must be met. If the charging response is considered, the EV power must reach the initial desired power before leaving the station under the premise of meeting the travel demand. If the discharge response is considered, the battery power requirement must also be met, and too low power will also affect the use of the EV, wherein the battery safety power constraint is:

[0193]

[0194] S max ≥S i (t) + S DR,i (t) ≥ S ex,i ;

[0195] S i (t) + S DR,i (t) ≥ S min ;

[0196] wherein, S minS i (t) represents the SOC of the user i of the electric vehicle at time t, S DR,i (t) represents the SOC of the user i of the electric vehicle at time t, S i (t) represents the SOC of the user i of the electric vehicle at time t, S max S represents the upper limit of the SOC of the battery of the electric vehicle, P d P represents the rated discharge power of the electric vehicle, P c P represents the rated charge power of the electric vehicle, t arrive t represents the arrival time of the user i of the electric vehicle, t leave t represents the expected departure time of the user i of the electric vehicle, η c C0 represents the charging efficiency when the electric vehicle is charging, C0 represents the capacity of the battery of the electric vehicle, S ex,i S represents the initial expected amount of electricity of the electric vehicle.

[0197] The higher the remaining amount of electricity of the EV when entering the station, the smaller the initial charging power, the larger the initial discharging power, and the lower the amount of electricity of the EV, the discharging power decreases, the higher the remaining amount of electricity of the EV when entering the station, the larger the initial charging power, the smaller the initial discharging power, and the higher the amount of electricity of the EV, the charging power decreases. The charge-discharge power constraint is:

[0198]

[0199] P min,i (t)≤P r,i (t)≤P max,i (t);

[0200] P c,i (t) represents the charging power, P r,i P represents the rated charge (discharge) power of the user i of the electric vehicle at time t, P d,i P represents the discharging power, P min,i P represents the minimum charge (discharge) power of the user i of the electric vehicle at time t, P max,i P represents the maximum charge (discharge) power of the user i of the electric vehicle at time t.

[0201] S342, determine the total response potential according to the charge-discharge response potential of the user of the electric vehicle, specifically:

[0202] P EV,i (t)=P nc,i (t)-P min,i (t);

[0203]

[0204] PEV,c,i (t) = P nc,i (t) ;

[0205] P EV,d,i (t) = η |P min,i (t) | ;

[0206] P EV,i (t) = P EV,c,i (t) + P EV,d,i (t) ;

[0207] P EV,min,i ≤ P EV,i ≤ P EV,max,i ;

[0208] In the formula, P EV,i (t) represents the charging and discharging response potential of the user i of the electric vehicle at time t, P nc,i (t) represents the natural charging power when not participating in the DR (demand response) process, P EV,c,i (t) represents the charging response potential of the user i of the electric vehicle at time t, P EV,d,i (t) represents the discharging response potential of the user i of the electric vehicle at time t, η represents, P EV,min,i represents the lower limit of the adjustable power of the electric vehicle i, P EV,max,i represents the upper limit of the adjustable power of the electric vehicle i, P EV,i represents the charging and discharging response potential of the user i of the electric vehicle.

[0209] S343, obtaining an electric vehicle cluster adjustable power calculation model according to the total response potential, the battery safety power constraint, and the charging and discharging power constraint, including:

[0210]

[0211] In the formula, ΔP EV (t) represents the adjustable power of the electric vehicle cluster at time t, and n represents the number of electric vehicles.

[0212] S4, aggregating the distributed power supply cluster adjustable power calculation model, the distributed energy storage cluster adjustable power calculation model, and the electric vehicle cluster adjustable power calculation model to obtain the adjustable range and the adjustable upper and lower limit constraints of the regional power distribution network flexibility resource.

[0213] In the face of multiple flexible resources of the distribution network, a unified model is needed to describe the flexibility characteristics of various resources and aggregate and equivalent them as a new whole, so that they can more efficiently participate in the interactive operation of the power market, and facilitate the unified management and dispatch of the power grid. Specifically, the adjustable power calculation model of the distributed power cluster, the adjustable power calculation model of the distributed energy storage cluster and the adjustable power calculation model of the electric vehicle cluster are aggregated using the Minkowski method to obtain the adjustable range and adjustable upper and lower limit constraints of the regional distribution network flexibility resource.

[0214] The adjustable range of the regional distribution network flexibility resource is:

[0215]

[0216] In the formula, Z all represents the adjustable range of the regional distribution network flexibility resource, Z j represents the adjustable range of the jth cluster, represents the Minkowski summation operation, ΔP PV (t) represents the adjustable power of the photovoltaic cluster at time t, ΔP W (t) represents the adjustable power of the wind power cluster at time t.

[0217] The power boundary represents the minimum (maximum) flexibility power that the cluster flexibility resource can provide at any time. The adjustable upper and lower limit constraints are:

[0218]

[0219] In the formula, P all,min represents the lower limit constraint of the distribution network adjustable power, P all,max represents the upper limit constraint of the distribution network adjustable power, i∈g represents the wind farm cluster, i∈w represents the photovoltaic power plant cluster, i∈l represents the energy storage cluster, and i∈n represents the electric vehicle cluster.

[0220] The present application considers various adjustable resources such as new energy generation, distributed energy storage and electric vehicles. New energy generation is renewable and widely distributed, and is distributed in various locations to reduce power transmission losses. However, its output is affected by weather conditions and time, and is volatile. Energy storage systems have very fast response times and can charge and discharge instantly, helping to smooth power grid fluctuations and respond to sudden demand, and can be configured as needed to provide different levels of capacity and power to meet specific grid regulation needs. Electric vehicles are equipped with large-capacity batteries, which can be considered as distributed energy and energy storage devices. Electric vehicles can be charged at different locations, and their mobility makes them a distributed resource in grid regulation.

[0221] The application groups distributed power sources, distributed energy storage and electric vehicles through a clustering algorithm, and establishes an adjustable power calculation model to analyze the power distribution network regulation capacity considering the above factors comprehensively.

[0222] Please refer to Figure 2 Embodiment two of the application is:

[0223] A power distribution network adjustable capacity evaluation terminal, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements each step of the above-mentioned power distribution network adjustable capacity evaluation method when executing the computer program.

[0224] To sum up, the application provides a power distribution network adjustable capacity evaluation method and terminal, which adopts a clustering algorithm to group and divide distributed power sources, distributed energy storage and electric vehicles based on wind power related data, photovoltaic related data, energy storage related data and electric vehicle related data, respectively, to obtain distributed power source clusters, distributed energy storage clusters and electric vehicle clusters, and establishes distributed power source cluster adjustable power calculation models, distributed energy storage cluster adjustable power calculation models and electric vehicle cluster adjustable power calculation models based on wind power related data, photovoltaic related data, energy storage related data and electric vehicle related data, respectively, aggregates the distributed power source cluster adjustable power calculation models, the distributed energy storage cluster adjustable power calculation models and the electric vehicle cluster adjustable power calculation models, and obtains the adjustable range and adjustable upper and lower limit constraints of the regional power distribution network flexibility resources. The application considers new energy generation, distributed energy storage and electric vehicles as multiple adjustable resources, uses a clustering algorithm combined with an adjustable power calculation model to evaluate the adjustable capacity of the regional power distribution network, thereby accurately evaluating the adjustable capacity of the power distribution network, helping to provide a basis and guidance for the participation of flexibility resources in peak shaving and frequency modulation, effectively reducing the peak-valley difference, stabilizing the load fluctuation and maintaining the stable operation of the power grid. Moreover, different clustering algorithms are used for cluster division of different flexibility resources, which is more suitable for the characteristics of flexibility resources, realizes reliable and accurate cluster division, and is convenient for subsequent accurate adjustable capacity evaluation. At the same time, different clustering algorithms are used for cluster division of different flexibility resources, which is more suitable for the characteristics of flexibility resources, realizes reliable and accurate cluster division, and is convenient for subsequent accurate adjustable capacity evaluation.

[0225] The above merely illustrates the embodiments of the present application, and is not intended to limit the patent scope of the present application. Any equivalent transformation or direct or indirect application in the related technical field based on the content of the present application specification and drawings is also included in the patent protection scope of the present application.

Claims

1. A method for evaluating the adjustability of a power distribution network, characterized in that, Including the following steps: Acquire data related to wind power, photovoltaics, energy storage, and electric vehicles in the power distribution network; Clustering algorithms are used to divide distributed power sources, distributed energy storage, and electric vehicles into clusters based on the wind power-related data, photovoltaic-related data, energy storage-related data, and electric vehicle-related data, respectively, to obtain distributed power source clusters, distributed energy storage clusters, and electric vehicle clusters; Based on the wind power related data, the photovoltaic related data, the energy storage related data, and the electric vehicle related data, respectively, adjustable power calculation models for the distributed power generation cluster, the distributed energy storage cluster, and the electric vehicle cluster are established. The adjustable power calculation models of the distributed power generation cluster, the distributed energy storage cluster, and the electric vehicle cluster are aggregated to obtain the adjustable range and upper and lower limits of the regional distribution network's flexibility resources. The clustering algorithm is used to divide distributed power sources, distributed energy storage, and electric vehicles into clusters based on the wind power-related data, photovoltaic-related data, energy storage-related data, and electric vehicle-related data, respectively, resulting in distributed power source clusters, distributed energy storage clusters, and electric vehicle clusters, including: The Fast Unfolding clustering algorithm is used to divide the distributed power sources into clusters based on the wind power-related data and the photovoltaic-related data, thus obtaining distributed power source clusters. The fuzzy C-means clustering algorithm is used to divide the distributed energy storage into clusters based on the energy storage-related data, resulting in distributed energy storage clusters; An improved k-means clustering algorithm was used to divide electric vehicles into clusters based on the electric vehicle-related data, thus obtaining electric vehicle clusters. The aggregation of the adjustable power calculation models of the distributed power generation cluster, the distributed energy storage cluster, and the electric vehicle cluster yields the adjustable range and upper and lower limits of the regional distribution network's flexibility resources, including: The Minkowski method is used to aggregate the adjustable power calculation models of the distributed power generation cluster, the distributed energy storage cluster, and the electric vehicle cluster to obtain the adjustable range and upper and lower limits of the regional distribution network's flexibility resources.

2. The method for evaluating the adjustability of a distribution network according to claim 1, characterized in that, The Fast Unfolding clustering algorithm is used to partition distributed power sources into clusters based on the wind power-related data and the photovoltaic-related data, resulting in distributed power source clusters including: The net load of the distributed power generation nodes is calculated based on the distributed power generation output and load of the distributed power generation nodes in the distribution network, and the net load of the distributed power generation nodes is used as the first grouping index. The reactive power regulation capacity and active power regulation capacity of distributed power sources are used as the second grouping indicators. The first cluster index and the second cluster index are normalized respectively to obtain the normalized first cluster index and the normalized second cluster index. The similarity between distributed power nodes is calculated based on the normalized first cluster index and the normalized second cluster index. A new cluster is obtained by randomly selecting the cluster containing any distributed power node and merging it with other clusters. The incremental value of the modularity function corresponding to the new cluster is calculated based on the similarity between the distributed power nodes. If the maximum increment value of the current modularity function is greater than the first preset value, then the cluster where any distributed power node is located is merged with the cluster corresponding to the maximum increment value of the modularity function to obtain a new cluster. If the maximum increment value of the current modularity function is not greater than the first preset value, then the cluster where any of the distributed power nodes is located will not be merged with the cluster corresponding to the maximum increment value of the modularity function. Treat the cluster as a new distributed power node, and return to the step of calculating the similarity between distributed power nodes based on the normalized first cluster index and the normalized second cluster index, until the maximum increment value of the modularity function no longer changes, thus obtaining a distributed power cluster.

3. The method for evaluating the adjustability of a distribution network according to claim 1, characterized in that, The improved k-means clustering algorithm is used to divide electric vehicles into clusters based on the electric vehicle-related data, resulting in electric vehicle clusters including: K electric vehicle nodes are randomly selected from the power distribution network as the initial cluster centers; Calculate the distances between the remaining electric vehicle nodes and the K cluster centers; The evaluation index of clustering effect is determined based on the universal gravitation model; Assign the remaining electric vehicle nodes corresponding to the maximum value of the evaluation index to the cluster center, and update the cluster center; Determine whether the evaluation index has reached the second preset value or the current iteration number has reached the preset number. If yes, stop the iteration and obtain the electric vehicle cluster. If no, return to the step of calculating the distance between the remaining electric vehicle nodes and the K cluster centers.

4. The method for evaluating the adjustability of a distribution network according to claim 1, characterized in that, The establishment of adjustable power calculation models for distributed power clusters, distributed energy storage clusters, and electric vehicle clusters based on the wind power-related data, photovoltaic-related data, energy storage-related data, and electric vehicle-related data respectively includes: Based on the wind power-related data, a wind power cluster adjustable power calculation model is established based on the distributed power cluster. Based on the photovoltaic-related data, a photovoltaic cluster adjustable power calculation model is established based on the distributed power generation cluster. Based on the energy storage-related data, an adjustable power calculation model for the distributed energy storage cluster is established. Based on the electric vehicle-related data, an adjustable power calculation model for the electric vehicle cluster is established.

5. The method for evaluating the adjustability of a distribution network according to claim 4, characterized in that, The step of establishing a wind power cluster adjustable power calculation model based on the distributed power cluster using the wind power-related data includes: The output power of the wind turbine is determined based on the aforementioned wind power-related data. The individual wind power adjustable potential is determined based on the output power of the wind turbine unit; A calculation model for the adjustable power of a wind power cluster is determined based on the adjustable potential of a single wind power source. The step of establishing a photovoltaic cluster adjustable power calculation model based on the distributed power cluster using the photovoltaic-related data includes: The output power of the photovoltaic unit is determined based on the aforementioned photovoltaic-related data; The adjustable potential of a single photovoltaic generator is determined based on the output power of the photovoltaic unit. The adjustable power calculation model for the photovoltaic cluster is determined based on the adjustable potential of the individual photovoltaic generator.

6. The method for evaluating the adjustability of a distribution network according to claim 4, characterized in that, The step of establishing an adjustable power calculation model for the distributed energy storage cluster based on the energy storage-related data includes: Establish energy storage charging and discharging constraints; The power of the energy storage load group is determined based on the energy storage-related data. Based on the power of the energy storage load group and the energy storage charging and discharging constraints, a calculation model for the adjustable power of the distributed energy storage cluster is obtained.

7. The method for evaluating the adjustability of a distribution network according to claim 4, characterized in that, The step of establishing an adjustable power calculation model for the electric vehicle cluster based on the electric vehicle-related data includes: Establish safety limits for electric vehicle batteries, including limits for charge and discharge power. The total response potential is determined based on the charge and discharge response potential of electric vehicle users; Based on the total response potential, the battery safety capacity constraint, and the charge / discharge power constraint, an adjustable power calculation model for electric vehicle clusters is obtained.

8. A terminal for assessing the adjustability of a power distribution network, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements each step of the method for evaluating the adjustability of a power distribution network according to any one of claims 1 to 7.

Images