A machine learning-supported energy storage configuration method and system for assisting large-scale new energy station access

Abstract

The present invention is applicable to the field of energy storage planning technology, and provides a method and system for energy storage configuration that assists access to large-scale new energy stations supported by machine learning, the method comprising: obtaining historical data on wind power and photovoltaic power generation output in a preset area; analyzing the historical data on wind power and photovoltaic power generation output using a machine learning algorithm to obtain a set of target output curves for the preset area; determining the scheduling constraints of daily cycle energy storage and the scheduling constraints of new energy stations respectively according to the typical output curve set; constructing an objective function for energy storage planning of a large-scale power system accessed by new energy stations, and determining an energy storage planning scheme based on the objective function, the scheduling constraints of daily cycle energy storage, and the scheduling constraints of new energy stations. The present invention improves the system's ability to absorb new energy, reduces wind and solar power abandonment, and optimizes the operating cost of the entire power system.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage planning, and in particular to a machine learning-supported energy storage configuration method and system for assisting access to large-scale new energy sites. Background Art

[0002] In the case of large-scale access to new energy, the volatility of new energy has a more obvious impact on the power grid. For example, with the substantial increase in wind power and photovoltaic installed capacity, the output power will change more in a short period of time. High penetration of new energy access requires the power grid to have higher flexibility. Since the power generation characteristics of new energy are different from those of traditional power sources such as thermal power, the power grid needs to be able to respond quickly to changes in new energy power. In addition, the new energy field itself also has a certain regulation capability. For example, some advanced wind turbines can adjust power output by changing the pitch angle, which can smooth power fluctuations to a certain extent. When configuring energy storage, it is necessary to fully consider how to combine with the self-regulation capability of new energy sites to form a complementary relationship.

[0003] There are limitations in the current energy storage configuration scheme. On the one hand, the penetration rate of new energy is at a low level. With the sharp increase in new energy installed capacity, the high proportion of grid connection has caused wind and solar power abandonment, sudden power changes, and grid frequency and voltage instability. On the other hand, it does not fully consider the self-regulation capabilities of new energy sites. For example, wind farms and photovoltaic power stations can actively control power but are ignored, resulting in isolated operation of energy storage and poor configuration. In addition, the interaction between sites and power grids is ignored. The existing energy storage strategy is difficult to meet the current large-scale access needs of new energy.

[0004] In view of this, there is a need for a machine learning-supported energy storage configuration method and system to assist in the access of large-scale new energy stations. Summary of the invention

[0005] The embodiments of the present application provide a method and system for energy storage configuration supported by machine learning to assist in accessing large-scale new energy sites, which is used to solve the problem that existing energy storage strategies are difficult to meet the current demand for large-scale access to new energy.

[0006] The first aspect of the embodiment of the present application provides a method for energy storage configuration supported by machine learning to assist access to large-scale new energy stations, including:

[0007] Obtain historical data on wind power and photovoltaic power generation output in a preset area;

[0008] Analyzing the historical data of wind power and photovoltaic power generation output using a machine learning algorithm to obtain a set of target output curves for the preset area;

[0009] Determining the scheduling constraints of the daily cycle energy storage and the scheduling constraints of the new energy station respectively according to the typical output curve set;

[0010] An objective function for energy storage planning of a large-scale power system connected to a new energy station is constructed, and an energy storage planning scheme is determined based on the objective function, the scheduling constraints of the daily cycle energy storage, and the scheduling constraints of the new energy station.

[0011] Furthermore, the historical data of wind power and photovoltaic power generation output are analyzed by using a machine learning algorithm to obtain a set of target output curves for the preset area, including:

[0012] Obtaining annual hourly output data of each new energy station in the preset area, and dividing the annual hourly output data into daily operation vectors;

[0013] Based on the divided daily operation vectors, cluster analysis is performed to obtain a set of typical output curves;

[0014] The typical day coefficient is calculated according to the typical output curve set.

[0015] Furthermore, the cluster analysis is performed based on the divided daily operation vectors to obtain a set of typical output curves, including:

[0016] Initialize cluster centers according to the preset number of clusters;

[0017] Calculate the Euclidean distance from each daily running vector to each cluster center;

[0018] Update the cluster center according to the calculated Euclidean distance result;

[0019] When the preset number of iterations is reached, the updated cluster center is the target typical output curve set.

[0020] Furthermore, updating the cluster center according to the calculated Euclidean distance result includes:

[0021]

[0022] Where: p′ cn is the updated cluster center, C n is the set of natural days belonging to the nth class, M is the set C n The number of natural days in p station,d is the daily running vector, It is the output of the new energy station ws in the i-th hour of the natural day d.

[0023] Furthermore, the scheduling constraints of the daily cycle energy storage and the scheduling constraints of the new energy stations include one or more of new energy consumption constraints, hourly dynamic characteristic constraints of new energy stations containing energy storage, and daily cycle energy storage operation constraints.

[0024] Furthermore, the expressions of the new energy consumption constraint, the hourly dynamic characteristic constraint of the new energy station with energy storage, and the daily cycle energy storage operation constraint are as follows:

[0025] Constraints on new energy consumption:

[0026]

[0027] Hourly dynamic characteristic constraints of new energy stations with energy storage:

[0028] (P ws,t +P ss,t )-(P ws,t-1 +P ss,t-1 )≤R ws

[0029] (P ws,t +P ss,t )-(P ws,t-1 +P ss,t-1 )≥-R ws

[0030] Daily cycle energy storage operation constraints:

[0031]

[0032] Soc ss,t ≤Cap ss

[0033] Soc ss,t =Soc ss,ini ,t=nT ss ,n=1,2,…

[0034] Among them, P ws,t and are the actual power consumption and output upper limit of the new energy field ws at time t, γ is the lower limit of consumption, which is a continuous rational number between 0 and 1, and P ss,t , Inject grid power, charging power and discharging power into the energy storage device, R ws is the station ws output change constraint, which is used to constrain the station output change range. ss,t is the charge state of the energy storage device at time t, Cap is the energy storage charging and discharging efficiency, ss is the energy storage capacity, and the energy storage return period is T ss Hour.

[0035] Furthermore, the objective function of the large-scale power system energy storage planning for the access of new energy stations is constructed, and the energy storage planning scheme is determined based on the objective function, the scheduling constraints of the daily cycle energy storage and the scheduling constraints of the new energy stations, including:

[0036] Objective function:

[0037]

[0038] Among them, τ p is the typical day coefficient, They are the variable cost, fixed cost, and start-up and shutdown cost of conventional units g; Cost of abandoned wind and solar power for new energy stations; P is the investment cost of energy storage equipment per unit capacity; g,t , They are the output of conventional units, abandoned wind and abandoned solar power; U g,t is the online state parameter of the conventional unit g, which is a 0-1 variable; V g,t Cap is the startup state parameter of the conventional unit g, which is a 0-1 variable; ss is the ss capacity of the energy storage equipment; P, G, W, S, and Stor are the typical day, conventional units, wind farms, photovoltaic farms, and the collection of energy storage equipment, respectively.

[0039] A second aspect of the embodiment of the present application provides a machine learning-supported energy storage configuration system for assisting access to large-scale new energy sites, including:

[0040] A historical data acquisition unit, used to acquire historical data of wind power and photovoltaic power generation output in a preset area;

[0041] a target output curve set determination unit, configured to analyze the historical data of the wind power and photovoltaic power generation outputs using a machine learning algorithm to obtain a target output curve set for the preset area;

[0042] A dispatch constraint determination unit, used to determine the dispatch constraints of the daily cycle energy storage and the dispatch constraints of the new energy station respectively according to the typical output curve set;

[0043] The energy storage planning scheme determination unit is used to construct an objective function for energy storage planning of a large-scale power system connected to a new energy site, and determine the energy storage planning scheme based on the objective function, the scheduling constraints of the daily cycle energy storage and the scheduling constraints of the new energy site.

[0044] A third aspect of the embodiments of the present application provides a computer device, including:

[0045] memories, transceivers, processors, and bus systems;

[0046] Wherein, the memory is used to store programs;

[0047] The processor is used to execute the program in the memory, including executing the energy storage configuration method for assisting the access of large-scale new energy stations supported by machine learning as described above;

[0048] The bus system is used to connect the memory and the processor so that the memory and the processor can communicate with each other.

[0049] A fourth aspect of an embodiment of the present application provides a readable storage medium, including instructions, which, when executed on a computer, enables the computer to execute the steps of the energy storage configuration method for assisting the access of large-scale new energy stations supported by machine learning as described above.

[0050] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0051] The present invention extracts a set of typical output curves of wind power and photovoltaic power by performing machine learning based on the historical data set of regional renewable energy output, and can analyze the output characteristics of renewable energy sites, providing an accurate data basis for subsequent energy storage configuration and scheduling. Through a reasonable scheduling model and energy storage planning, it is achieved that when large-scale renewable energy is connected to the system, the combination of renewable energy sites and energy storage can be flexibly adjusted according to the actual needs of the power grid, effectively smoothing the intermittent and volatility of renewable energy, ensuring the power balance and stable operation of the power system, and at the same time greatly improving the system's ability to absorb renewable energy, reducing the phenomenon of wind and solar power abandonment, and optimizing the operating cost of the entire power system.

[0052] Other advantages, objectives, and features of the present invention will be set forth in part in the following description, and in part will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A schematic flow chart of an embodiment of a method for energy storage configuration supported by machine learning to assist access to large-scale new energy stations in the present invention;

[0054] Figure 2 The present invention is a flowchart of another embodiment of a method for energy storage configuration supported by machine learning to assist access to large-scale new energy stations. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0056] In this embodiment, the energy storage configuration method supported by machine learning to assist large-scale new energy stations in accessing is used to optimize the operating cost of the power system. In this embodiment, the implementation method can be implemented in the system, on the server, or on the terminal, without specific limitation.

[0057] Embodiment 1

[0058] See also Figure 1 , an embodiment of a method for energy storage configuration supported by machine learning to assist large-scale new energy station access comprises the following steps:

[0059] S11. Obtain historical data of wind power and photovoltaic power generation output in a preset area;

[0060] In this embodiment, the preset region is any set region, and the historical data of wind power and photovoltaic power generation output can be obtained from the internal data recording system of the new energy station, the power dispatching department or the energy regulatory department. After the data is obtained, data preprocessing is required, where data preprocessing includes missing value and outlier processing, unifying the data time resolution, and normalizing the data; classifying the data type by station, marking it by time, and performing data seasonal analysis, for example, analyzing the wind power and photovoltaic power generation power change trends in different seasons, and providing basic data for subsequent typical output curve extraction and energy storage scheduling model.

[0061] S12. Analyze historical data of wind power and photovoltaic power generation output using a machine learning algorithm to obtain a set of typical output curves for a preset area;

[0062] In this embodiment, please refer to Figure 2 , step S12 further includes the following steps:

[0063] S121. Obtain the annual hourly output data of each new energy station in the preset area, and divide the annual hourly output data into daily operation vectors;

[0064] Specifically, annual hourly output data is obtained from each renewable energy station in the region. These data should be able to represent the output of renewable energy stations in the region at different times. The data source can be the monitoring system of the station or the database of the power dispatching department. To ensure the integrity and accuracy of the data, the data is preliminarily checked and cleaned to remove data points with obvious errors or missing data points. The preprocessing method is similar to the above steps. In addition, the annual hourly output data of each renewable energy station is divided into daily operation vectors p station,d , for each natural day d, in It is the output of the new energy station ws in the i-th hour of the natural day d.

[0065] S122. Perform cluster analysis based on the divided daily operation vectors to obtain a set of typical output curves;

[0066] Step S122 also includes the following:

[0067] 1. Initialize the cluster center according to the preset number of clusters;

[0068] N cluster centers are randomly generated, where N is a preset number of clusters, which can be selected based on experience or data characteristics. For example, N can be determined by roughly estimating the number of possible typical output modes through preliminary observation of the data.

[0069] 2. Calculate the Euclidean distance from each daily running vector to each cluster center;

[0070] For each daily run vector p station,d , calculate it to each cluster center p cn The Euclidean distance dist(d,c n ). According to the principle of minimum Euclidean distance, each natural day d is classified into the typical day category represented by the corresponding cluster center. That is, if dist(d,c k ) is the smallest distance, then the natural day d belongs to the distance p ck The clustering centered on .

[0071] 3. Update the cluster center according to the calculated Euclidean distance result;

[0072] Based on the classification results of the previous step, the average daily running vector of each natural day in each category is calculated and used as the new cluster center. The specific calculation formula is:

[0073]

[0074] Where: p′ cn is the updated cluster center, C n is the set of natural days belonging to the nth class, M is the set C n The number of natural days in .

[0075] 4. When the preset number of iterations is reached, the updated cluster center is the target typical output curve set.

[0076] Repeat the above steps of calculating the Euclidean distance, classifying, and updating the cluster center until the cluster center is no longer updated or the pre-set number of iterations is reached. In each iteration, the classification of natural days is continuously adjusted so that the daily operation vectors in each category are more similar and the cluster center can better represent the typical output mode of the category.

[0077] S123. Calculate the typical day coefficient based on the typical output curve set.

[0078] When the clustering process converges, the daily operation vector represented by each cluster center is the set of typical output curves of the current station. These typical output curves can reflect the output change law of the new energy station under different typical working conditions.

[0079] For each typical output curve p b , count the number of natural days corresponding to it, this number is the typical day coefficient The typical day coefficient can reflect the frequency of each typical output mode in the data set, which is helpful to further analyze the output characteristics of new energy stations.

[0080] S13. Determine the dispatch constraints of the daily cycle energy storage and the dispatch constraints of the new energy station respectively according to the typical output curve set;

[0081] In this embodiment, the scheduling constraints of daily cycle energy storage and the scheduling constraints of new energy stations include one or more of new energy consumption constraints, hourly dynamic characteristics constraints of new energy stations containing energy storage, and daily cycle energy storage operation constraints.

[0082] Constraints on new energy consumption:

[0083]

[0084] Hourly dynamic characteristic constraints of new energy stations with energy storage:

[0085] (P ws,t +P ss,t )-(P ws,t-1 +P ss,t-1 )≤R ws

[0086] (P ws,t +P ss,t )-(P ws,t-1 +P ss,t-1 )≥-R ws

[0087] Daily cycle energy storage operation constraints:

[0088]

[0089] Soc ss,t ≤Cap ss

[0090] Soc ss,t =Soc ss,ini ,t=nT ss ,n=1,2,…

[0091] Among them, P ws,t and are the actual power consumption and output upper limit of the new energy field ws at time t, γ is the lower limit of consumption, which is a continuous rational number between 0 and 1, and P ss,t , Inject grid power, charging power and discharging power into the energy storage device, R ws is the station ws output change constraint, which is used to constrain the station output change range. ss,t is the charge state of the energy storage device at time t, Cap is the energy storage charging and discharging efficiency, ss is the energy storage capacity, and the energy storage return period is T ss Hour.

[0092] S14. Construct an objective function for energy storage planning of a large-scale power system connected to new energy stations, and determine an energy storage planning scheme based on the objective function, the scheduling constraints of daily cycle energy storage, and the scheduling constraints of new energy stations.

[0093] In this embodiment, the objective function of the energy storage planning model for a large-scale power system including a new energy station, wherein the objective function takes the minimum comprehensive cost as the target parameter, is specifically expressed as follows:

[0094]

[0095] Among them, τ p is the typical day coefficient, They are the variable cost, fixed cost, and start-up and shutdown cost of conventional units g; Cost of abandoned wind and solar power for new energy stations; P is the investment cost of energy storage equipment per unit capacity; g,t , They are the output of conventional units, abandoned wind and abandoned solar power; U g,t is the online state parameter of the conventional unit g, which is a 0-1 variable; V g,t Cap is the startup state parameter of the conventional unit g, which is a 0-1 variable; ss is the ss capacity of the energy storage equipment; P, G, W, S, and Stor are the typical day, conventional units, wind farms, photovoltaic farms, and the collection of energy storage equipment, respectively.

[0096] Finally, the above objective function is optimized and solved, and the operating characteristics and costs of conventional units, new energy stations and energy storage equipment in the power grid are comprehensively considered to provide decision support for energy storage planning of large-scale power systems, so as to reasonably allocate energy storage resources, improve new energy absorption capacity and reduce system operating costs.

[0097] Embodiment 2

[0098] An embodiment of a machine learning-supported energy storage configuration system for assisting large-scale new energy station access in the present invention includes the following steps:

[0099] A historical data acquisition unit, used to acquire historical data of wind power and photovoltaic power generation output in a preset area;

[0100] A target output curve set determination unit is used to analyze historical data of wind power and photovoltaic power generation output using a machine learning algorithm to obtain a target output curve set for a preset area;

[0101] A dispatch constraint determination unit, used to determine the dispatch constraints of daily cycle energy storage and the dispatch constraints of new energy stations respectively according to a set of typical output curves;

[0102] The energy storage planning scheme determination unit is used to construct the objective function of the energy storage planning of the large-scale power system connected to the new energy site, and determine the energy storage planning scheme based on the objective function, the scheduling constraints of the daily cycle energy storage and the scheduling constraints of the new energy site.

[0103] For the specific definition of the system, please refer to the definition of the method above, which will not be repeated here. Each module in the above system can be implemented in whole or in part by software, hardware and a combination thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0104] Embodiment 3

[0105] The present invention provides a computer device, comprising a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, the steps of the above method are implemented.

[0106] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0107] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored. In addition, each functional unit in each embodiment of the present invention can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units.

[0108] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes 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, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0109] It can be understood 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 above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A machine learning-supported energy storage configuration method for assisting large-scale new energy station access, characterized in that: include: Obtain historical data on wind power and photovoltaic power generation output in a preset area; Analyzing the historical data of wind power and photovoltaic power generation output using a machine learning algorithm to obtain a set of target output curves for the preset area; Determining the scheduling constraints of the daily cycle energy storage and the scheduling constraints of the new energy station respectively according to the typical output curve set; An objective function for energy storage planning of a large-scale power system connected to a new energy station is constructed, and an energy storage planning scheme is determined based on the objective function, the scheduling constraints of the daily cycle energy storage, and the scheduling constraints of the new energy station.

2. The energy storage configuration method for assisting large-scale new energy station access supported by machine learning according to claim 1 is characterized in that: The method of analyzing the historical data of wind power and photovoltaic power generation output by using a machine learning algorithm to obtain a target output curve set for the preset area includes: Obtaining annual hourly output data of each new energy station in the preset area, and dividing the annual hourly output data into daily operation vectors; Based on the divided daily operation vectors, cluster analysis is performed to obtain a set of typical output curves; The typical day coefficient is calculated according to the typical output curve set.

3. The energy storage configuration method for assisting large-scale new energy station access supported by machine learning according to claim 2 is characterized in that: The cluster analysis is performed based on the divided daily operation vectors to obtain a set of typical output curves, including: Initialize cluster centers according to the preset number of clusters; Calculate the Euclidean distance from each daily running vector to each cluster center; Update the cluster center according to the calculated Euclidean distance result; When the preset number of iterations is reached, the updated cluster center is the target typical output curve set.

4. The energy storage configuration method for assisting large-scale new energy station access supported by machine learning according to claim 3 is characterized in that: The updating of the cluster center according to the calculated Euclidean distance result includes: Where: p′ cn is the updated cluster center, C n is the set of natural days belonging to the nth class, M is the set C n The number of natural days in p station,d is the daily running vector, It is the output of the new energy station ws in the i-th hour of the natural day d.

5. The energy storage configuration method supported by machine learning to assist large-scale new energy station access according to claim 1 is characterized in that: The dispatching constraints of the daily cycle energy storage and the dispatching constraints of the new energy stations include one or more of new energy consumption constraints, hourly dynamic characteristic constraints of the new energy stations containing energy storage, and daily cycle energy storage operation constraints.

6. The energy storage configuration method for assisting large-scale new energy station access supported by machine learning according to claim 5 is characterized in that: The expressions of the new energy consumption constraint, the hourly dynamic characteristic constraint of the new energy station with energy storage, and the daily cycle energy storage operation constraint are as follows: Constraints on new energy consumption: Hourly dynamic characteristic constraints of new energy stations with energy storage: (P ws,t +P ss,t )-(P ws,t-1 +P ss,t-1 )≤R ws (P ws,t +P ss,t )-(P ws,t-1 +P ss,t-1 )≥-R ws Daily cycle energy storage operation constraints: Soc ss,t ≤Cap ss Soc ss,t =Soc ss,ini ,t=nT ss ,n=1,2,… Among them, P ws,t and are the actual power consumption and output upper limit of the new energy field ws at time t, γ is the lower limit of consumption, which is a continuous rational number between 0 and 1, and P ss,t , Inject grid power, charging power and discharging power into the energy storage device, R ws is the station ws output change constraint, which is used to constrain the station output change range. ss,t is the charge state of the energy storage device at time t, Cap is the energy storage charging and discharging efficiency, ss is the energy storage capacity, and the energy storage return period is T ss Hour.

7. The energy storage configuration method for assisting large-scale new energy station access supported by machine learning according to claim 1 is characterized in that: The objective function of the large-scale power system energy storage planning for the access of new energy stations is constructed, and the energy storage planning scheme is determined based on the objective function, the scheduling constraints of the daily cycle energy storage and the scheduling constraints of the new energy stations, including: Objective function: Among them, τ p is the typical day coefficient, They are the variable cost, fixed cost, and start-up and shutdown cost of conventional units g; The cost of abandoning wind and solar power at new energy stations; P is the investment cost of energy storage equipment per unit capacity; g,t , They are the output of conventional units, abandoned wind and abandoned solar power; U g,t is the online state parameter of the conventional unit g, which is a 0-1 variable; V g,t Cap is the startup state parameter of the conventional unit g, which is a 0-1 variable; ss is the ss capacity of the energy storage equipment; P, G, W, S, and Stor are the typical day, conventional units, wind farms, photovoltaic farms, and the collection of energy storage equipment, respectively.

8. A machine learning-supported energy storage configuration system for assisting large-scale new energy stations to access, characterized in that: include: A historical data acquisition unit, used to acquire historical data of wind power and photovoltaic power generation output in a preset area; a target output curve set determination unit, configured to analyze the historical data of the wind power and photovoltaic power generation outputs using a machine learning algorithm to obtain a target output curve set for the preset area; A dispatch constraint determination unit, used to determine the dispatch constraints of the daily cycle energy storage and the dispatch constraints of the new energy station respectively according to the typical output curve set; The energy storage planning scheme determination unit is used to construct an objective function for energy storage planning of a large-scale power system connected to a new energy site, and determine the energy storage planning scheme based on the objective function, the scheduling constraints of the daily cycle energy storage and the scheduling constraints of the new energy site.

9. A computer device, characterized in that: include: memories, transceivers, processors, and bus systems; Wherein, the memory is used to store programs; The processor is used to execute the program in the memory, including executing the energy storage configuration method for assisting the access of large-scale new energy stations supported by machine learning according to any one of claims 1 to 7; The bus system is used to connect the memory and the processor so that the memory and the processor can communicate with each other.

10. A readable storage medium, characterized in that: It includes instructions, which, when executed on a computer, enable the computer to execute the steps of the energy storage configuration method for assisting the access of large-scale new energy stations supported by machine learning as described in any one of claims 1 to 7.

Images