Distributed energy storage aggregation method and system based on improved Minkowski sum

Abstract

The invention discloses a distributed energy storage aggregation method and system based on the improved Minkowski sum, and relates to the technical field of power systems, and the method comprises the steps: constructing a single operation model; according to the operation model of each energy storage monomer, representing a power regulation feasible region of the energy storage monomer in a Polytopes semi-plane form; clustering the to-be-aggregated distributed energy storage cluster according to the model parameter of each distributed energy storage monomer in the to-be-aggregated distributed energy storage cluster, and determining a clustering result of the to-be-aggregated distributed energy storage cluster; and according to clustering results, aggregating the power regulation feasible regions of all monomer energy storage in each type of distributed energy storage by adopting an improved internal approximate Minkowski sum, and determining an aggregation feasible region of each clustering result of the energy storage cluster to be aggregated. According to the method, the aggregation efficiency of a large number of distributed flexible resources is improved, the flexibility lost in the aggregation process is reduced, the adjusting capacity of the power system is enhanced, and positive support is provided for sustainable development and economic operation of the power system.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a distributed energy storage aggregation method and system based on an improved Minkowski sum. Background Art

[0002] With the continuous advancement of the electric energy substitution strategy, the power load has increased year by year. This trend has brought unprecedented challenges to the safety and economic operation of the power system. In the face of increasingly severe energy demand, it is particularly important to effectively tap and mobilize distributed energy storage resources on the user side to participate in power peak regulation. This can not only reduce dependence on traditional power investment and reduce the peak-to-valley difference of the power system, but also significantly improve the operational safety and economy of the power system. Although the adjustable power of single energy storage equipment is usually small and it is difficult to play a significant role in large-scale peak regulation, the distributed energy storage group has unique advantages: large number, flexible scheduling methods and huge potential. Therefore, how to effectively aggregate the adjustable capacity of distributed energy storage resources and achieve efficient scheduling and management has become a technical problem that needs to be solved urgently in the power system. Summary of the invention

[0003] In view of the above-mentioned problems, the present invention is proposed.

[0004] Therefore, the technical problem solved by the present invention is: how to solve the problems of difficulty in coordinating and scheduling multiple energy storage units in the existing distributed energy storage system, inaccurate power regulation, and mutual interference between clusters.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a distributed energy storage aggregation method based on improved Minkowski sum, which comprises the following steps:

[0006] Collect parameter data based on distributed energy storage on the user side;

[0007] Construct a single-unit operation model based on the operating principle of distributed energy storage;

[0008] According to the operation model of each energy storage unit, the power regulation feasible domain is represented in the form of a half-plane of Polytopes;

[0009] Clustering the distributed energy storage clusters to be aggregated according to the model parameters of each distributed energy storage unit in the distributed energy storage clusters to be aggregated, and determining the clustering results of the distributed energy storage clusters to be aggregated;

[0010] According to the clustering results, the improved inner approximate Minkowski sum is used to aggregate the power regulation feasible domains of all single energy storages in each type of distributed energy storage, and the aggregated feasible domain of each clustering result of the energy storage cluster to be aggregated is determined.

[0011] As a preferred solution of the distributed energy storage aggregation method based on improved Minkowski sum described in the present invention, the parameter data includes the capacity, rated power, charging and discharging efficiency, and initial SOC of each distributed energy storage.

[0012] As a preferred solution of the distributed energy storage aggregation method based on the improved Minkowski sum described in the present invention, wherein: the monomer operation model is expressed as:

[0013]

[0014] in, is the energy storage level of energy storage device n at time t; and are charging efficiency and discharging efficiency respectively; and are the charging and discharging power of energy storage device n at time t respectively; and are the minimum and maximum values ​​of the charging power of the energy storage device n, respectively; and are the minimum and maximum values ​​of the discharge power of the energy storage device n, respectively; and are the minimum capacity and maximum capacity of energy storage device n respectively; B ESSc,t and B ESSdis,t It is the charge and discharge state variable, which is a 0-1 variable.

[0015] As a preferred solution of the distributed energy storage aggregation method based on the improved Minkowski sum described in the present invention, wherein: the characterization of the power regulation feasible domain in the form of a half-plane of Polytopes is to characterize the power feasible domain of the energy storage device n in the half-plane of Polytopes, and the expression is:

[0016]

[0017] in, and are the power feasible domain of energy storage device n and the power matrix in the scheduling period; and are the inequality coefficient matrix and column vector representing the feasible domain of energy storage device n respectively; represents the power constraint, represents the power constraint; E is the unit matrix, Determined by the charge and discharge efficiency.

[0018] As a preferred solution of the distributed energy storage aggregation method based on the improved Minkowski sum described in the present invention, wherein: the distributed energy storage clusters to be aggregated are clustered according to the model parameters of each distributed energy storage monomer in the distributed energy storage clusters to be aggregated, and the clustering results of the distributed energy storage clusters to be aggregated are determined, including constructing a data sample set according to the model parameters of each monomer distributed energy storage in the distributed energy storage clusters to be aggregated, and normalizing the initial data; setting the neighborhood radius r and the minimum density M min ; Randomly select a non-clustered point O 1 according to Computing and O 1 The number of points whose Euclidean distance is less than r is M l , and O 1 Gather into one category;

[0019] Among them, D E (x1,x2) is the Euclidean distance between vectors x1 and x2; x 1,i and x 2,i is the i-th dimension of vectors x1 and x2; I is the total number of dimensions.

[0020] As a preferred solution of the distributed energy storage aggregation method based on the improved Minkowski sum described in the present invention, wherein: the clustering result of determining the distributed energy storage cluster to be aggregated also includes:

[0021] When M l ≥M min When , traverse the other points in the current class, calculate the number of unclustered points in the neighborhood r of each point, and include the unclustered points in the current class. If the number of unclustered points in the neighborhood of all points in the current class is less than M min , then re-judge whether there are still clustered points and continue to calculate and classify. Otherwise, re-traverse other points in the current class and calculate the number of non-clustered points in the neighborhood r of each point again;

[0022] When M l <M min When , it is determined whether there are still cluster points, and the number of calculations M is continued. l And classify.

[0023] As a preferred solution of the distributed energy storage aggregation method based on the improved Minkowski sum described in the present invention, wherein: the improved inner approximate Minkowski sum is used to aggregate the power regulation feasible domains of all monomer energy storages in each type of distributed energy storage, and the aggregation feasible domain of each clustering result of the energy storage cluster to be aggregated is determined to include:

[0024] According to the operation model of energy storage, the original feasible domain of single energy storage is obtained Select parameters and perform DBSCAN clustering to form multiple energy storage clusters;

[0025] Calculate initial benchmarks for each cluster and Solve the optimization problem to obtain the feasible domain of each single energy storage in the cluster after approximation The expression is:

[0026]

[0027] Get the aggregate feasible domain of the current cluster According to calculating and recording the accuracy index of each cluster, the expression is:

[0028]

[0029] Change the A0 of each cluster, perform inner approximation again and record the accuracy index. After traversing the A0 of each cluster, the feasible domain of cluster aggregation with the highest final accuracy is formed.

[0030] Another object of the present invention is to provide a distributed energy storage aggregation system based on an improved Minkowski sum, which can collect the operating parameters of energy storage monomers, construct a polyhedral power regulation model, classify different monomer energy storage devices based on a clustering algorithm, and use an improved inner approximate Minkowski sum algorithm to achieve optimized aggregation of the power feasible domain, effectively solving the problems of scheduling incoordination, insufficient power regulation accuracy, and poor system scalability and robustness caused by large differences in energy storage device parameters in the prior art.

[0031] In order to solve the above technical problems, the present invention provides the following technical solutions: a distributed energy storage aggregation system based on improved Minkowski sum, comprising: a data collection and preprocessing module, a monomer operation model building module, a cluster analysis module, a Minkowski sum aggregation and optimization module;

[0032] The data collection and preprocessing module collects parameter data from the distributed energy storage device on the user side, and cleans and standardizes the collected data;

[0033] The monomer operation model building module is to establish a mathematical model for each energy storage monomer according to the operating principle of distributed energy storage;

[0034] The cluster analysis module automatically clusters the energy storage devices based on the model parameters and operating status of each energy storage unit, uses the DBSCAN clustering algorithm to classify energy storage units with similar characteristics into the same cluster, sets the neighborhood radius and minimum density of the clustering algorithm, and completes the adaptive classification;

[0035] The Minkowski sum aggregation and optimization module uses an improved inner approximate Minkowski sum algorithm to aggregate the power regulation feasible domain of each cluster to form an aggregated feasible domain of the cluster, calculates the accuracy index multiple times, performs inner approximate optimization, and finally achieves the optimal accuracy of the aggregated feasible domain.

[0036] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the distributed energy storage aggregation method based on the improved Minkowski sum are implemented as described above.

[0037] A computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, the steps of the distributed energy storage aggregation method based on the improved Minkowski sum are implemented as described above.

[0038] Beneficial effects of the invention: The invention aims at the unprecedented challenges brought by the high proportion of new energy grid connection and the increasing power load year by year to the safety and economic operation of the power system, and proposes a distributed energy storage cluster aggregation method based on the improved inner approximate Minkowski sum. Compared with the prior art, the invention improves the aggregation efficiency of a large number of distributed flexible resources, reduces the flexibility lost in the aggregation process, enhances the regulation ability of the power system, and provides active support for the sustainable development and economic operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0040] Figure 1 An overall flow chart of a distributed energy storage aggregation method based on an improved Minkowski sum provided for the first embodiment of the present invention;

[0041] Figure 2 A schematic diagram of distributed energy storage clustering results in a distributed energy storage aggregation method based on an improved Minkowski sum provided in a second embodiment of the present invention;

[0042] Figure 3 A schematic diagram of power and capacity aggregation results of a distributed energy storage cluster 1 based on a distributed energy storage aggregation method based on an improved Minkowski sum provided in a second embodiment of the present invention;

[0043] Figure 4A schematic diagram of power and capacity aggregation results of a distributed energy storage cluster 2 based on a distributed energy storage aggregation method based on an improved Minkowski sum is provided in accordance with a second embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0045] Example 1, reference Figure 1 As an embodiment of the present invention, a distributed energy storage aggregation method based on improved Minkowski sum is provided, comprising:

[0046] Step 1: According to the operating principle of distributed energy storage, the single-unit operation model is constructed as follows:

[0047]

[0048]

[0049] In the formula, is the energy storage level of energy storage device n at time t; and are charging efficiency and discharging efficiency respectively; and are the charging and discharging power of energy storage device n at time t respectively; and are the minimum and maximum values ​​of the charging power of the energy storage device n, respectively; and are the minimum and maximum values ​​of the discharge power of the energy storage device n, respectively; and are the minimum capacity and maximum capacity of energy storage device n respectively; B ESSc,t and B ESSdis,t It is the charge and discharge state variable, which is a 0-1 variable.

[0050] Step 2: The power regulation feasible domain of each energy storage unit’s operating model is represented in the form of a half-plane of Polytopes:

[0051]

[0052] In the formula, and are the power feasible domain of energy storage device n and the power matrix in the scheduling period; and are the inequality coefficient matrix and column vector representing the feasible domain of energy storage device n respectively; represents the power constraint, represents the power constraint; E is the unit matrix, Determined by the charge and discharge efficiency; the specific meaning is as follows:

[0053]

[0054]

[0055] Step 3: clustering the distributed energy storage clusters to be aggregated according to the model parameters of each distributed energy storage unit in the distributed energy storage clusters to be aggregated, and determining the clustering results of the distributed energy storage clusters to be aggregated, including:

[0056] Step 1: construct a data sample set according to the model parameters of each single distributed energy storage in the distributed energy storage cluster to be aggregated, and normalize the initial data;

[0057] Step 2: Set the neighborhood radius r and minimum density M min ;

[0058] Step 3: Randomly select an unclustered point O 1 according to Computing and O 1 The number of points whose Euclidean distance is less than r is M l , and these points and O 1 Clustered into one category, among which: D E (x1,x2) is the Euclidean distance between vectors x1 and x2; x 1,i and x 2,i is the i-th dimension of vectors x1 and x2; I is the total number of dimensions.

[0059] Step 4: If M l ≥M min Go to step 5, otherwise go to step 7;

[0060] Step 5: Traverse other points in the current class, calculate the number of unclustered points in the neighborhood r of each point, and include these unclustered points in the current class;

[0061] Step 6: If the number of unclustered points in the neighborhood of all points in the current class is less than M min , go to step 7 otherwise return to step 5;

[0062] Step 7: If there are still unclustered points, return to step 3, otherwise clustering is completed.

[0063] Step 4: Use the improved inner approximate Minkowski sum to aggregate the power regulation feasible domains of all single energy storages in each type of distributed energy storage, including:

[0064] The power regulation feasible domain of all single energy storage in each type of distributed energy storage cluster is approximated internally:

[0065]

[0066] In the formula, is the approximate constraint space of individual n; σ n is the scaling factor of individual n; is the translation factor of individual n, and its specific value can be obtained by solving the following LP problem:

[0067]

[0068] In the formula, α n =1 / σ n ; N is the number of distributed energy storage in this type of cluster.

[0069] The Minkowski sum is performed on the approximate power regulation feasible domains of all single energy storages in each type of distributed energy storage cluster to determine the aggregated feasible domain of each clustering result of the energy storage group to be aggregated:

[0070]

[0071] In the formula, Ω agg It is the power aggregation feasible domain of distributed energy storage cluster.

[0072] In order to reduce the flexibility lost in the inner approximation process, the following accuracy index is defined, based on which the approximation degree is improved by changing the reference A0:

[0073]

[0074] Where N f is the number of unit normal vectors; Δ C,l and Δ A,l are the widths of the original feasible domain and the approximate feasible domain in the direction of the lth unit normal vector, respectively.

[0075] Accuracy index calculation:

[0076] Assume Δ O is the maximum distance between two tangent planes in the original feasible domain in the direction of each unit normal vector, which can be solved by optimization method. The original feasible O is represented by a half-space polyhedron as follows:

[0077] O={x∈R N :Ax≤b} (16)

[0078] For a unit normal vector f l =[a1,a2,…,a N ], and its corresponding N-dimensional hyperplane is:

[0079] a1x1+a2x2+…+a N a N +b l =0 (17)

[0080] In the formula, x1, x2, …, x N is the position in each dimension; b l Is a constant.

[0081] Assumptions:

[0082] X P =[x 1p ,x 2p ,…,x Np ] (18)

[0083] X Q =[x 1q ,x 2q ,…,x Nq ] (19)

[0084] Where, X P and X Q They represent a point outside the hyperplane and a point inside the hyperplane respectively.

[0085] The unit normal vector f l Δ on O,l , which can be solved according to the following optimization model:

[0086]

[0087] In summary, the design is based on the improved inner approximate Minkowski and feasible domain aggregation method as follows:

[0088]

[0089] So far, a distributed energy storage cluster aggregation model based on the improved inner approximate Minkowski sum has been established and can be applied to actual examples.

[0090] Therefore, according to the aggregated feasible domain of each clustering result of the determined distributed energy storage cluster to be aggregated, the adjustable capacity of the distributed energy storage cluster to be aggregated can be evaluated. In addition, the aggregated feasible domain of the distributed energy storage cluster to be aggregated can be submitted to the power grid dispatching center for dispatching strategy.

[0091] Embodiment 2 is an embodiment of the present invention, which provides a system for a distributed energy storage aggregation method based on an improved Minkowski sum, including: a data collection and preprocessing module, a monomer operation model building module, a cluster analysis module, a Minkowski sum aggregation and optimization module;

[0092] The data collection and preprocessing module collects parameter data from the distributed energy storage devices on the user side and cleans and standardizes the collected data;

[0093] The single-unit operation model building module builds a mathematical model for each energy storage unit based on the operating principle of distributed energy storage;

[0094] The clustering analysis module automatically clusters energy storage devices based on the model parameters and operating status of each energy storage unit. It uses the DBSCAN clustering algorithm to classify energy storage units with similar characteristics into the same cluster, sets the neighborhood radius and minimum density of the clustering algorithm, and completes adaptive classification.

[0095] The Minkowski sum aggregation and optimization module uses an improved inner approximate Minkowski sum algorithm to aggregate the feasible domains of individual power regulation within each cluster to form the cluster's aggregated feasible domain, calculates accuracy indicators multiple times, performs inner approximate optimization, and ultimately achieves the optimal level of accuracy for the aggregated feasible domain.

[0096] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program code.

[0097] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0098] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0099] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0100] Embodiment 3: In this embodiment, in order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments. This embodiment experiments the existing traditional method and the method of this embodiment respectively.

[0101] The present invention aggregates 50 distributed energy storages, and the parameter distribution range is shown in Table 1.

[0102] Table 1 Distribution of distributed energy storage parameters

[0103]

[0104] Table 2 shows the calculation results of the accuracy index before and after improvement, where the accuracy index of distributed energy storage cluster 1 before improvement was 85.41%, and the accuracy index after improvement was 90.57%, with an accuracy improvement of 5.16%; the accuracy index of distributed energy storage cluster 2 before improvement was 95.80%, and the accuracy index after improvement was 98.25%, with an accuracy improvement of 2.45%; it can be seen that the method proposed in the present invention can effectively reduce the flexibility lost in the approximation process and improve the degree of approximation.

[0105] Table 2 Calculation results of accuracy index

[0106]

[0107] Figure 2 This is the clustering result of distributed energy storage. Cluster parameter 1 is selected as the power / power ratio, which reflects the time required for the energy storage to be fully charged from the rated power. Cluster parameter 2 is selected as the charging and discharging efficiency, which determines the shape of the feasible domain A matrix. The number of clusters 1 is 43, and the number of clusters 2 is 7;

[0108] Figure 3 The power and capacity aggregation results of distributed energy storage cluster 1 are shown in Figure 1. According to the aggregation results, the upper limit of charging power of cluster 1 is 6323.84kW, the lower limit of discharging power is -6315.55kW, the upper limit of capacity is 15471.47kWh, and the lower limit of capacity is -3726.65kWh. It can be seen that distributed energy storage has strong charging and discharging capabilities and energy storage capabilities after aggregation, and can absorb a large amount of energy during charging so as to release it during peak grid demand.

[0109] Figure 4 This is the power and capacity aggregation result of distributed energy storage cluster 2. According to the aggregation result, the upper limit of charging power of aggregation cluster 2 is 197.92kW, the lower limit of discharging power is -197.72kW, the upper limit of capacity is 1110.39kWh, and the lower limit of capacity is -274.42kWh.

[0110] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A distributed energy storage aggregation method based on improved Minkowski sum, characterized in that: include: Collect parameter data based on distributed energy storage on the user side; Construct a single-unit operation model based on the operating principle of distributed energy storage; According to the operation model of each energy storage unit, the power regulation feasible domain is represented in the form of a half-plane of Polytopes; Clustering the distributed energy storage clusters to be aggregated according to the model parameters of each distributed energy storage unit in the distributed energy storage clusters to be aggregated, and determining the clustering results of the distributed energy storage clusters to be aggregated; According to the clustering results, the improved inner approximate Minkowski sum is used to aggregate the power regulation feasible domains of all single energy storages in each type of distributed energy storage, and the aggregated feasible domain of each clustering result of the energy storage cluster to be aggregated is determined.

2. A distributed energy storage aggregation method based on improved Minkowski sum as claimed in claim 1, characterized in that: The parameter data includes the capacity, rated power, charging and discharging efficiency, and initial SOC of each distributed energy storage.

3. A distributed energy storage aggregation method based on improved Minkowski sum as claimed in claim 2, characterized in that: The monomer operation model is expressed as: in, is the energy storage level of energy storage device n at time t; and are charging efficiency and discharging efficiency respectively; and are the charging and discharging power of energy storage device n at time t respectively; and are the minimum and maximum values ​​of the charging power of the energy storage device n, respectively; and are the minimum and maximum values ​​of the discharge power of the energy storage device n, respectively; and are the minimum capacity and maximum capacity of energy storage device n respectively; B ESSc,t and B ESSdis,t It is the charge and discharge state variable, which is a 0-1 variable.

4. A distributed energy storage aggregation method based on improved Minkowski sum as claimed in claim 3, characterized in that: The method of characterizing the power regulation feasible region in the form of a half-plane of Polytopes is to characterize the power feasible region of the energy storage device n in the form of a half-plane of Polytopes, and the expression is: in, and are the power feasible domain of energy storage device n and the power matrix in the scheduling period; and are the inequality coefficient matrix and column vector representing the feasible domain of energy storage device n respectively; represents the power constraint, represents the power constraint; E is the unit matrix, Determined by the charge and discharge efficiency.

5. A distributed energy storage aggregation method based on improved Minkowski sum as claimed in claim 4, characterized in that: The step of clustering the distributed energy storage clusters to be aggregated according to the model parameters of each distributed energy storage unit in the distributed energy storage clusters to be aggregated and determining the clustering results of the distributed energy storage clusters to be aggregated includes constructing a data sample set according to the model parameters of each distributed energy storage unit in the distributed energy storage clusters to be aggregated, and normalizing the initial data; Set the neighborhood radius r and minimum density M min ; Randomly select a non-clustered point O 1 according to Computing and O 1 The number of points whose Euclidean distance is less than r is M l , and O 1 Gather into one category; Among them, D E (x1,x2) is the Euclidean distance between vectors x1 and x2; x 1,i and x 2,i is the i-th dimension of vectors x1 and x2; I is the total number of dimensions.

6. A distributed energy storage aggregation method based on improved Minkowski sum as claimed in claim 5, characterized in that: The step of determining the clustering result of the distributed energy storage cluster to be aggregated further includes: When M l ≥M min When , traverse the other points in the current class, calculate the number of unclustered points in the neighborhood r of each point, and include the unclustered points in the current class. If the number of unclustered points in the neighborhood of all points in the current class is less than M min , then re-judge whether there are still clustered points and continue to calculate and classify. Otherwise, re-traverse other points in the current class and calculate the number of non-clustered points in the neighborhood r of each point again; When M l <M min When , it is determined whether there are still cluster points, and the number of calculations M is continued. l And classify.

7. A distributed energy storage aggregation method based on improved Minkowski sum as claimed in claim 6, characterized in that: The improved inner approximate Minkowski sum is used to aggregate the power regulation feasible domains of all single energy storages in each type of distributed energy storage, and the aggregated feasible domain of each clustering result of the energy storage cluster to be aggregated is determined to include: According to the operation model of energy storage, the original feasible domain of single energy storage is obtained Select parameters and perform DBSCAN clustering to form multiple energy storage clusters; Calculate initial benchmarks for each cluster and Solve the optimization problem to obtain the feasible domain of each single energy storage in the cluster after approximation The expression is: Get the aggregate feasible domain of the current cluster According to calculating and recording the accuracy index of each cluster, the expression is: Change the A0 of each cluster, perform inner approximation again and record the accuracy index. After traversing the A0 of each cluster, the feasible domain of cluster aggregation with the highest final accuracy is formed.

8. A system using a distributed energy storage aggregation method based on improved Minkowski sum as claimed in any one of claims 1 to 7, characterized in that: It includes data collection and preprocessing module, monomer operation model building module, cluster analysis module, Minkowski and aggregation and optimization module; The data collection and preprocessing module collects parameter data from the distributed energy storage device on the user side, and cleans and standardizes the collected data; The monomer operation model building module is to establish a mathematical model for each energy storage monomer according to the operating principle of distributed energy storage; The cluster analysis module automatically clusters the energy storage devices based on the model parameters and operating status of each energy storage unit, uses the DBSCAN clustering algorithm to classify energy storage units with similar characteristics into the same cluster, sets the neighborhood radius and minimum density of the clustering algorithm, and completes the adaptive classification; The Minkowski sum aggregation and optimization module uses an improved inner approximate Minkowski sum algorithm to aggregate the power regulation feasible domain of each cluster to form an aggregated feasible domain of the cluster, calculates the accuracy index multiple times, performs inner approximate optimization, and finally achieves the optimal accuracy of the aggregated feasible domain.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a distributed energy storage aggregation method based on an improved Minkowski sum according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a distributed energy storage aggregation method based on improved Minkowski sum as described in any one of claims 1 to 7 are implemented.

Images