Charging and discharging prediction method and system for regional energy storage

Abstract

The invention provides a charging and discharging prediction method and system for regional energy storage, relates to the field of energy storage electric quantity management, and solves the technical problems that a set of regional aggregation modes capable of dynamically adapting is difficult to set according to the characteristics of regional energy storage, and dynamic registration of energy storage regions is difficult to realize in supply and demand distribution. According to the method, planning is performed based on released electric quantity and class received electric quantity to obtain an energy storage area; predicting the predicted power consumption of each energy storage area in the day where the target time is located, and dividing each energy storage area into a first area, a second area and a third area based on the predicted power consumption and the maximum storage power; a target second area of the first area is identified, an electric quantity adjusting instruction corresponding to the first area is determined based on the under-electric quantity of the target second area, the real-time charging power of the energy storage device in the first area is determined, and discharging corresponding to the first area is performed based on the real-time storage electric quantity of the target second area; according to the invention, the intelligent collaboration and scheduling capability of the energy storage resources in the regional range can be improved.

Description

Technical Field

[0001] This application relates to the field of energy storage power management, and in particular to a method and system for predicting charge and discharge for regional energy storage. Background Technology

[0002] With the continuous increase in renewable energy penetration and the accelerated advancement of energy transition, electrochemical energy storage, as a key technology supporting peak shaving and valley filling in the power grid, smoothing out fluctuations in new energy sources, and improving power supply reliability, is rapidly developing towards large-scale, grid-based, and regional deployment. In regional energy storage systems, a large number of distributed energy storage devices are scattered throughout the power grid, and their operating status is closely coupled with power flow, electricity price signals, and local loads.

[0003] However, existing regional energy storage management technologies face two major challenges: First, in terms of regional division, they often rely on fixed administrative geographical boundaries or simple physical distances, ignoring the actual "roles" played by each energy storage device in the power grid and their dynamic behavior patterns. This results in poor functional coordination within the divided energy storage areas, making it difficult to form an efficient scheduling synergy. Second, in terms of supply and demand scheduling, traditional forecasting and allocation methods are often fragmented. Forecasting models are easily affected by abnormal data, while supply and demand matching often fails to comprehensively consider power adaptability, line losses, and system complexity, making it difficult for scheduling schemes to achieve optimal economic efficiency and real-time performance. These two problems further make it difficult to achieve accurate load forecasting and efficient cross-regional energy collaborative scheduling during the implementation of regional energy storage charging and discharging, and it is difficult to improve the overall operational efficiency of regional energy storage systems by achieving dynamic registration of energy storage areas in supply and demand allocation.

[0004] Therefore, this invention discloses a charging and discharging prediction method and system for regional energy storage, which is used to solve the above-mentioned technical problems. Summary of the Invention

[0005] This application provides a charging and discharging prediction method and system for regional energy storage, which solves the technical problems in the prior art of charging and discharging prediction for regional energy storage, namely, the difficulty in setting a dynamically adaptable regional aggregation mode according to the characteristics of regional energy storage, and the difficulty in achieving dynamic registration of energy storage regions in supply and demand allocation.

[0006] To achieve the above objectives, this application adopts the following technical solution: Firstly, a method for predicting charge and discharge for regional energy storage is provided, including: S1: Obtain the released and received power of each energy storage device the day before the target time, and plan the energy storage devices based on the released and received power to obtain the energy storage area; where the target time is manually set and can be 0:00:01 of each day; the energy storage device is a manually set basic energy storage unit, specifically an energy storage device that can be uniformly managed for charging and discharging, such as an independent energy storage cabinet or multiple energy storage cabinets connected together; the received power is the power received by the power supply area corresponding to the energy storage device from the energy storage power supply of other areas; S2: Based on historical electricity consumption, predict the predicted electricity consumption of each energy storage area on the day of the target time. Divide each energy storage area into a first area, a second area, and a third area based on the predicted electricity consumption and the maximum stored capacity. Identify the target second area of ​​the first area. Determine the corresponding power adjustment command for the first area based on the insufficient power of the target second area. Determine the real-time charging power of the energy storage device in the first area based on the power adjustment command. Discharge the corresponding first area based on the real-time stored capacity of the target second area.

[0007] In conjunction with the first aspect above, in one possible implementation, the energy storage area is obtained by planning the energy storage device based on the released electricity and the received electricity, including: Release of electricity from each energy storage device and class receive power The standard value is obtained by calculating formula (1). and standard value This step releases electrical energy. and class receive power The range is mapped to [0,1]; based on standard values and standard value The stability of net power outflow is determined by formula (2). Through standard values Standard value and stability of net power outflow Constructing 3D feature vectors ;in, The serial number of the energy storage device; Based on the three-dimensional feature vector corresponding to the energy storage device The number is determined by formula (3). The energy storage device and its number are Behavioral similarity between energy storage devices Among them, the storage number is The three-dimensional feature vectors corresponding to the energy storage devices are respectively Number The three-dimensional feature vector corresponding to the energy storage device is , The serial number of the energy storage device; Obtain the latitude and longitude of each energy storage device, and determine its number based on the latitude and longitude using formula (4). The energy storage device and its number are Geographic proximity between energy storage devices ; behavioral similarity Geographic proximity The synergistic value is obtained by adding the weights together. The weights are added together as follows: , It is a proportional adjustment coefficient determined based on the manager's preference for behavioral roles and geographical proximity, and The value range is (0.1, 0.9), when When the settings are large, the system prioritizes grouping devices with highly consistent behavioral roles into the same area, even if they are slightly far apart; when When the hour is set, the system places more emphasis on geographical proximity and tends to aggregate devices that are physically close together; Based on collaborative value Identify energy storage areas; The calculation formula (1) is: ; In the formula, To release electricity Standardized value For receiving electricity The standardized value; The maximum value of the released electricity from all energy storage devices. This represents the minimum amount of electricity released by all energy storage devices. The maximum value of the received electricity among all energy storage devices. This is the minimum value of the received electricity among all energy storage devices. The calculation formula (2) is: ; In the formula, To prevent extremely small positive numbers from being divided by zero, it is obtained through manual setting; The calculation formula (3) is: ; The calculation formula (4) is: ; In the formula, For the number The energy storage device and its number are The Euclidean distance between energy storage devices is determined by latitude and longitude. It is a scale parameter used to control the rate at which proximity decays with distance, and is determined based on the geographical distance between all energy storage devices.

[0008] In conjunction with the first aspect mentioned above, one possible implementation method is based on collaborative value. Determine the energy storage area, including: A1: Treating each energy storage device as a node, if the collaborative value between two nodes... If the value is greater than the collaborative value threshold, a virtual edge is established between the two nodes; wherein the weight of the virtual edge is the corresponding collaborative value. The value; the synergy value threshold is determined by analyzing all synergy values. Choose a collaborative value that can filter out the bottom 20% or 30% of the values. As a threshold for collaborative value; A2: Traverse all nodes and mark nodes with more than a certain number of virtual edges as core nodes; where the number threshold is set by the number of energy storage devices. A3: Mark the core node that is not marked as a seed point and has the largest number of virtual edges as a region. Seed points; among them, It is the area code; A4: Find all regions any internal node Non-core nodes that are connected by virtual edges and have not yet been classified are determined by formula (5) to identify several of the aforementioned non-core nodes and regions. Merging attraction ; A5: Select the merge attraction. Median maximum merger attraction If the value is the largest merging attraction If the value exceeds the attraction threshold, the corresponding node will be assigned to the region. Repeat steps A4 and A5 until no new nodes can be added to the region. Stop the current area. The establishment of the attraction threshold is based on historical aggregated attraction. It is confirmed; A6: Determine if there are any core nodes that are not marked as seed points; if yes, jump to A3; if no, proceed to the constructed region. Mark them as independent energy storage areas and include those not planned for the area The energy storage devices are marked as independent energy storage areas; The calculation formula (5) is: ; In the formula, For the non-core nodes and regions Each node The sum of their synergistic values area The current number of nodes.

[0009] In conjunction with the first aspect above, one possible implementation involves predicting the electricity consumption of each energy storage area on the day containing the target time, based on historical electricity consumption data. This includes: The charging time is extracted from the regional database, and the non-charging time of the day is marked as the discharge time; the charging time is obtained by setting the historical electricity price for each time period in the n days prior to the target time. The electricity consumption of each energy storage area within the dischargeable time n days prior to the target time is obtained from the regional database; the electricity consumption of several areas is integrated into a power consumption group, the variance of the electricity consumption within the power consumption group is obtained, and it is determined whether the variance is greater than the power variance threshold value; where n is obtained by seasonal setting. If so, mark the average value of the electricity consumption in the electricity consumption group as the reference electricity consumption, remove the electricity consumption in the electricity consumption group with the largest absolute value of the difference from the reference electricity consumption, recalculate the variance of the remaining electricity consumption in the electricity consumption group and re-perform the variance judgment until the variance of the electricity consumption group is less than the variance threshold value, and calculate the average value of the remaining electricity consumption in the electricity consumption group to obtain the characteristic electricity consumption corresponding to the day of the target time. If not, the characteristic electricity consumption corresponding to the day of the target time is obtained by averaging the electricity consumption within the electricity consumption group; wherein, the variance threshold is automatically determined based on the statistical distribution of historical electricity consumption data. Obtain the characteristic electricity consumption and actual electricity consumption corresponding to the dischargeable time for n days prior to the target time. Obtain the ratio of actual electricity consumption to characteristic electricity consumption for each day. Calculate the characteristic ratio by averaging several ratios. Multiply the characteristic ratio by the characteristic electricity consumption corresponding to the day of the target time to obtain the predicted electricity consumption corresponding to the dischargeable time for the day of the target time. Wherein, the ratio of actual electricity consumption to characteristic electricity consumption for each day is the actual electricity consumption for each day divided by the characteristic electricity consumption for the corresponding day.

[0010] In conjunction with the first aspect mentioned above, in one possible implementation, the energy storage areas are divided into a first area, a second area, and a third area based on predicted electricity consumption and maximum storage capacity, including: The maximum stored power is obtained by adding up the maximum stored power of each energy storage device in the energy storage area, and the predicted power consumption corresponding to each energy storage area is extracted. When the maximum stored power of the energy storage area is greater than the predicted power consumption, the energy storage area is marked as the first area; when the maximum stored power of the energy storage area is less than the predicted power consumption, the energy storage area is marked as the second area; when the maximum stored power of the energy storage area is equal to the predicted power consumption, the energy storage area is marked as the third area.

[0011] In conjunction with the first aspect described above, in one possible implementation, identifying the target second region of the first region includes: B1: Subtract the predicted power consumption from the maximum stored power in the first region to obtain the surplus power in the first region. The predicted power consumption of the second region is subtracted from the maximum stored power to obtain the power deficit of the second region. ;in, This is the number of the first region. The numbering of the second area; B2: Based on surplus power and low battery The power supply compatibility of each first and second region is determined by formula (6). ; B3: Extract the number from the regional database. The first area and numbered Power loss rate between the second region Based on power loss rate and power supply compatibility The number is determined by formula (7). The first area and numbered Power supply potential index between the second region Among them, the power loss rate is based on the numbered The first area and numbered The line length and line impedance between the second region are calculated. B4: Based on surplus power Sort the first region in descending order of quantity; B5: Extract the first region sequentially based on the sorting number from smallest to largest, and then extract the power supply potential index corresponding to the first region. The index numbers are obtained by sorting them from largest to smallest. The index numbers are then updated after removing the index numbers corresponding to the target second region that have been marked as other first regions. B6: Extract the index numbers in ascending order based on the updated index number, and determine whether the current index number is greater than the index number threshold; if so, mark the second region corresponding to the index number before the index number threshold as the target second region of the current first region, and jump to B5 to traverse the remaining first regions; where the index number threshold is a threshold for the number of second regions to avoid a first region supplying too much power. If not, compare the current index number with the corresponding under-charge of the index numbers preceding the current index number. The total deficit is obtained by dividing by the corresponding power loss difference rate and then adding them together; when the total deficit is less than the current surplus power of the first region... Repeat step B6; when the total under-charge is not less than the current surplus charge in the first region. At that time, mark the second region corresponding to the index number before the current index number as the target second region of the current first region, and jump to B5 to traverse the remaining first regions; where the power loss difference rate is one minus the power deficit. The power loss rate corresponding to the second region and the current first region The value of, i.e. The value; The calculation formula (6) is: ; The calculation formula (7) is: ; In the formula, The baseline loss rate is used to convert different units of measurement. After normalization, the power loss rate between each first region and the second region can be obtained. The average value; and It is the proportional adjustment coefficient, and It can be dynamically adjusted according to the system operation strategy.

[0012] In conjunction with the first aspect above, in one possible implementation, determining the power adjustment command corresponding to the first region based on the insufficient power in the target second region includes: Extract the target second region from the first region, and determine the low power of the target second region. The total deficit power of the target second region is obtained by dividing by the corresponding power loss difference rate and then adding them together; when the total deficit power is equal to the surplus power... When the ratio exceeds the ratio threshold one, a power adjustment command one is issued; among them, the ratio threshold one, ratio threshold two and ratio threshold three are the key parameters for classifying the system response level, and are ordered from largest to smallest as ratio threshold one, ratio threshold two and ratio threshold three, and their values ​​can be optimized according to the system operation strategy and historical data. When the total short power and the surplus power When the ratio exceeds the ratio threshold 2 but does not exceed the ratio threshold 1, power adjustment command 2 is issued; When the total short power and the surplus power When the ratio does not exceed the ratio threshold two, power adjustment command three is issued.

[0013] In conjunction with the first aspect above, in one possible implementation, determining the real-time charging power of the energy storage device within the first region based on the power adjustment command includes: When the power adjustment command is power adjustment command one, the energy storage device in the first area is charged at maximum power, and charging stops when the power stored in the energy storage device in the first area is greater than the sum of the current predicted power consumption in the first area and the total power deficit in the corresponding target second area. When the power adjustment command is power adjustment command two, the energy storage devices in the first region are charged at maximum power before the reference time point within the charging time, and at adaptive power after the reference time point within the charging time. Charging stops when the stored power in the energy storage devices in the first region exceeds the sum of the predicted power consumption of the first region and the total shortfall in the corresponding target second region. The reference time point is manually set and is generally located at 70% of the charging time. The specific calculation method for adaptive power charging is as follows: the sum of the predicted power consumption of the first region and the total shortfall in the corresponding target second region is marked as the required power; the required power is subtracted from the total existing power of the energy storage devices in the first region to obtain the region's uncharged amount; the storable power of each energy storage device in the first region is divided by the total storable power of the first region to obtain the power allocation ratio; the region's uncharged amount is multiplied by the allocation ratio to obtain the uncharged amount of each energy storage device; the duration between the current time and the end time of the charging time is marked as the charging duration; the uncharged amount is divided by the charging duration to obtain the adaptive power charging for each energy storage device. When the power adjustment command is power adjustment command three, the energy storage device in the first area is charged with adaptive power, and charging stops when the power stored in the energy storage device in the first area is greater than the sum of the current predicted power consumption of the first area and the total power deficit of the corresponding target second area.

[0014] In conjunction with the first aspect mentioned above, in one possible implementation, discharging the corresponding first region based on the real-time stored energy in the target second region includes: The target second regions in the first region are sorted from largest to smallest amount of electricity deficit, and based on the sorting, electricity equal to the amount of electricity deficit is supplied to the target second regions through the first region in sequence.

[0015] Secondly, a charging and discharging prediction system for regional energy storage is provided, comprising: a regional planning module and a linked energy storage module; the regional planning module is used to obtain the released power and similar received power of each energy storage device on the day before the target time, and to plan the energy storage devices based on the released power and similar received power to obtain the energy storage area; the similar received power is the power received by the power supply area corresponding to the energy storage device from the energy storage power supply of other areas. The linked energy storage module is used to predict the predicted electricity consumption of each energy storage area on the day of the target time based on historical electricity consumption, divide each energy storage area into a first area, a second area, and a third area based on the predicted electricity consumption and the maximum stored capacity; identify the target second area of ​​the first area, determine the corresponding power adjustment command for the first area based on the insufficient power of the target second area, determine the real-time charging power of the energy storage device in the first area based on the power adjustment command, and discharge the corresponding first area based on the real-time stored capacity of the target second area.

[0016] This application provides a method and system for predicting charge and discharge for regional energy storage, with the following advantages: 1. This invention obtains the released and received power of each energy storage device on the day before the target time, and plans the energy storage devices based on the released and received power to obtain energy storage areas; predicts the predicted power consumption of each energy storage area on the day of the target time based on historical power consumption, and divides each energy storage area into a first area, a second area, and a third area based on the predicted power consumption and the maximum stored power; identifies the target second area of ​​the first area, determines the power adjustment command for the corresponding first area based on the insufficient power of the target second area, determines the real-time charging power of the energy storage devices in the first area based on the power adjustment command, and discharges the corresponding first area based on the real-time stored power of the target second area. This solves the technical problems of existing technologies in predicting the charging and discharging of regional energy storage, which make it difficult to set a dynamically adaptable regional aggregation mode according to the characteristics of regional energy storage, and difficult to achieve dynamic registration of energy storage areas in supply and demand allocation; this invention can improve the intelligent coordination and efficient scheduling capability of energy storage resources across the entire region.

[0017] 2. The energy storage region generated by this invention exhibits high synergy among its members in both behavior and space. This greatly simplifies the complexity of subsequent prediction and linkage control. The upper-level linkage energy storage module no longer needs to deal with a large number of discrete and independently behaving units, but interacts with several "regional groups" with clear objectives and internal coordination. This enables more accurate prediction of the overall regional electricity demand and more efficient execution of cross-regional power dispatch and charging / discharging strategies. Ultimately, this achieves optimized energy allocation at the entire power grid level, enhances the capacity for renewable energy absorption, and strengthens the stability and economy of power grid operation.

[0018] 3. This method transforms the complex problem of energy storage device relationships into a clear problem of network community discovery. The process is highly automated, robust to noise, and can adapt to network structure. The divided regions have strong internal coordination and clear boundaries between regions. This not only greatly improves the scientificity and efficiency of regional planning, but also directly empowers subsequent charging and discharging prediction and coordinated scheduling. It is a key technological breakthrough for realizing refined and intelligent management of regional energy storage.

[0019] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0020] Figure 1 A schematic diagram illustrating the steps of a charge / discharge prediction method for regional energy storage provided in an embodiment of this application; Figure 2 A schematic diagram illustrating the steps for planning and obtaining an energy storage area provided in this application embodiment; Figure 3 This is a schematic diagram of a charge and discharge prediction system for regional energy storage provided in an embodiment of this application. Detailed Implementation

[0021] In the description of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" means one or more, and "multiple" means two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences.

[0022] It should be noted that, in this application, the terms "exemplary" or "for example" are used to indicate that something is being described as an example, illustration, or illustration. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0023] like Figure 1 As shown in the embodiment of this application, a charge / discharge prediction method for regional energy storage includes: S1: Obtain the released and received power of each energy storage device the day before the target time, and plan the energy storage devices based on the released and received power to obtain the energy storage area; where the target time is manually set and can be 0:00:01 of each day; the energy storage device is a manually set basic energy storage unit, specifically an energy storage device that can be uniformly managed for charging and discharging, such as an independent energy storage cabinet or multiple energy storage cabinets connected together; the received power is the power received by the power supply area corresponding to the energy storage device from the energy storage power supply of other areas; S2: Based on historical electricity consumption, predict the predicted electricity consumption of each energy storage area on the day of the target time. Divide each energy storage area into a first area, a second area, and a third area based on the predicted electricity consumption and the maximum stored capacity. Identify the target second area of ​​the first area. Determine the corresponding power adjustment command for the first area based on the insufficient power of the target second area. Determine the real-time charging power of the energy storage device in the first area based on the power adjustment command. Discharge the corresponding first area based on the real-time stored capacity of the target second area.

[0024] It is worth noting that this invention, through regional dynamic planning, adaptive linkage, and refined control mechanisms, fundamentally solves the core pain points of existing regional energy storage systems, such as closed-loop forecasting, unreasonable resource allocation, and low energy utilization efficiency, achieving intelligent coordination and efficient scheduling of energy storage resources across the entire region. First, by deeply analyzing the previous day's released and received electricity through the regional planning module, energy storage areas can be scientifically divided. This planning method, based on actual operating data, better aligns with the actual power flow distribution and energy exchange characteristics of the power grid compared to static or simple load-based methods, laying a solid foundation for subsequent precise control and effectively avoiding localized energy storage idleness or overload problems caused by improper regional division. Second, the linked energy storage analysis step dynamically divides the region into three zones: a first zone, a second zone, and a third zone. This three-dimensional classification method makes the power grid's supply and demand situation readily apparent, greatly simplifying decision-making complexity and providing clear guidance for the directional flow of energy. Meanwhile, this invention establishes a complete closed-loop control chain from identification and decision-making to execution. By accurately identifying the target support area in the surplus region and generating refined power adjustment instructions based on the real-time power shortage in the target area, it realizes the transformation from "blind discharge" to "on-demand supply." This precise "point-to-point" energy support not only minimizes energy loss during long-distance transmission but also significantly improves the grid's absorption capacity and operational stability. In summary, this invention constructs an adaptive and highly efficient regional energy storage operation system through data-driven regional planning, intelligent regional classification, and precise linkage control. Its widespread application will greatly improve the grid integration and absorption level of new energy sources, enhance the flexibility and resilience of the grid, and provide strong technical support for achieving a clean energy structure transformation, possessing significant economic value and strategic significance.

[0025] In one possible implementation of this application embodiment, the above-mentioned S1 can be implemented by the following S101 and S102, which are described in detail below: S101: Based on the released electricity and the received electricity, the energy storage area is planned to include: Release of electricity from each energy storage device and class receive power The standard value is obtained by calculating formula (1). and standard value This step releases electrical energy. and class receive power The range is mapped to [0,1]; based on standard values and standard value The stability of net power outflow is determined by formula (2). Through standard values Standard value and stability of net power outflow Constructing 3D feature vectors ;in, The energy storage device is numbered; in this step, the stability of net power outflow is measured. This is used to quantify the role tendency of energy storage devices in the power grid, and its value range is [-1, 1]. When the value is close to 1, it indicates that the energy storage device mainly functions as a discharge unit; when... When the value is close to -1, it indicates that the energy storage device primarily functions as a charging unit; when... When the value is close to 0, it indicates that the charging and discharging behavior of the device is relatively balanced; introducing As the third dimension of the three-dimensional feature vector, it enables the calculation of behavioral similarity. At that time, not only the magnitude of charging and discharging was considered. and Furthermore, role orientation was taken into consideration, which enabled the more accurate grouping of devices with "similar roles" together, thereby enhancing the functionality and purposefulness of the area planning. Based on the three-dimensional feature vector corresponding to the energy storage device The number is determined by formula (3). The energy storage device and its number are Behavioral similarity between energy storage devices Among them, the storage number is The three-dimensional feature vectors corresponding to the energy storage devices are respectively Number The three-dimensional feature vector corresponding to the energy storage device is , This step is used to measure the consistency of the direction of the feature vectors of two energy storage devices, which can ensure that devices with "similar roles" are grouped together, such as two strong discharge power sources or two strong charging holes. The range is [ [1,1], the closer the value is to 1, the more similar the behavioral roles of the two devices are, such as both being net discharge power sources or net charge holes; the closer the value is to -1, the more completely opposite the roles are, such as one discharging and the other charging; the closer the value is to 0, the more balanced one role is, or the two are not closely related. Obtain the latitude and longitude of each energy storage device, and determine its number based on the latitude and longitude using formula (4). The energy storage device and its number are Geographic proximity between energy storage devices This step is numbered as follows: The energy storage device and its number are The closer the energy storage devices are, the greater the Euclidean distance. The smaller, The closer to 1; the farther away, The closer to 0; behavioral similarity Geographic proximity The synergistic value is obtained by adding the weights together. The weights are added together as follows: , It is a proportional adjustment coefficient determined based on the manager's preference for behavioral roles and geographical proximity, and The value range is (0.1, 0.9), when When the settings are large, the system prioritizes grouping devices with highly consistent behavioral roles into the same area, even if they are slightly far apart; when When the hour is set, the system places more emphasis on geographical proximity and tends to aggregate devices that are physically close together; Based on collaborative value Identify energy storage areas; The calculation formula (1) is: ; In the formula, To release electricity Standardized value For receiving electricity The standardized value; The maximum value of the released electricity from all energy storage devices. This represents the minimum amount of electricity released by all energy storage devices. The maximum value of the received electricity among all energy storage devices. This is the minimum value of the received electricity among all energy storage devices. The calculation formula (2) is: ; In the formula, To prevent extremely small positive numbers from being divided by zero, it is obtained through manual setting; The calculation formula (3) is: ; The calculation formula (4) is: ; In the formula, For the number The energy storage device and its number are The Euclidean distance between energy storage devices is determined by latitude and longitude. It is a scale parameter used to control the rate at which proximity decays with distance, and is determined based on the geographical distance between all energy storage devices.

[0026] It is worth noting that the method of planning energy storage devices based on the released power and the similar received power to obtain energy storage areas in this invention, through refined data modeling, multi-dimensional fusion evaluation and flexible strategy control, jointly constructs a regional energy storage planning system that is both scientific and forward-looking. First, this method introduces the indicator of "net power outflow stability" and constructs a three-dimensional feature vector together with standardized released power and similar received power. This transcends the limitations of traditional methods that group energy storage devices based solely on single power values ​​or simple geographical locations. By quantifying the "role orientation" of energy storage devices—as a discharge source, charging hole, or balancing unit—it profoundly reveals their inherent attributes in the power grid function. This makes the subsequent clustering process no longer a simple physical or numerical proximity, but a high degree of aggregation of behavioral patterns and functional roles, thus significantly improving the functional homogeneity and collaborative operation efficiency of the formed energy storage areas. For example, grouping devices that all play a strong discharge role can better cope with regional peak loads, while grouping devices that all play a strong charging role can more efficiently absorb surplus energy. This precise role-based classification directly lays a solid foundation for advanced applications of the power grid such as peak shaving and valley filling, and emergency backup. Second, this method achieves an organic integration of behavioral similarity and geographical proximity. It does not separate or simply superimpose the two, but constructs an adjustable weighting coefficient. The synergistic value function empowers managers to make flexible decisions based on actual grid topology, line capacity loss, control costs, and strategic objectives. When functional synergy is emphasized, it can enhance… Weighting ensures optimal role matching for maximum scheduling efficiency; when emphasizing physical access and line investment costs, it can reduce... The weighting of data, prioritizing geographical proximity to reduce line losses and construction costs, employs a flexible trade-off mechanism that allows regional planning schemes to accurately adapt to diverse application scenarios and management preferences, achieving a dynamic optimization balance between economic benefits and power grid safety performance. Furthermore, the standardization process employed in this method ensures the robustness of data processing and the accuracy of model evaluation, eliminates interference from data of different dimensions on the analysis results, effectively captures the directional consistency of feature vectors, and smoothly reflects the impact of geographical distance, avoiding abrupt changes caused by rigid threshold divisions. This appropriate use of mathematical tools guarantees the scientific rigor and stability of the clustering results. Ultimately, the energy storage region generated by this method exhibits high levels of coordination among its members in both behavior and space. This significantly simplifies the complexity of subsequent prediction and coordinated control. The upper-level coordinated energy storage module no longer needs to deal with a large number of discrete, independently behaving units, but instead interacts with several "regional clusters" with clear objectives and internal coordination. This enables more accurate prediction of the overall regional electricity demand and more efficient execution of cross-regional power dispatch and charging / discharging strategies. Ultimately, this achieves optimized energy allocation at the entire power grid level, enhances the absorption capacity of renewable energy, and strengthens the stability and economy of power grid operation. This demonstrates the significant progress and immense application value of this method in promoting the intelligent and refined management of regional energy storage.

[0027] S102: Based on Collaborative Value Determine the energy storage area, including: A1: Treating each energy storage device as a node, if the collaborative value between two nodes... If the value is greater than the collaborative value threshold, a virtual edge is established between the two nodes; wherein the weight of the virtual edge is the corresponding collaborative value. The value; the synergy value threshold is determined by analyzing all synergy values. Choose a collaborative value that can filter out the bottom 20% or 30% of the values. As a threshold for collaborative value; the bottom represents the collaborative value at the end after sorting from largest to smallest. In this step, the virtual edge represents whether there is a "meaningful cooperative relationship" between two energy storage devices. If the two energy storage devices are far apart and their behaviors are completely unrelated, then the cooperative value is negligible. The value will be very low. To prevent such "weak ties" from interfering with the judgment, a threshold is used to "filter" noise, leaving only those with collaborative value. Only a sufficiently high level of relationship is recognized as a "virtual edge"; A2: Traverse all nodes and mark nodes with more than one virtual edge as core nodes; the number of virtual edges is set by the number of energy storage devices and can be 5. A3: Mark the core node that is not marked as a seed point and has the largest number of virtual edges as a region. Seed points; among them, It is the area code; A4: Find all regions any internal node Non-core nodes that are connected by virtual edges and have not yet been classified are determined by formula (5) to identify several of the aforementioned non-core nodes and regions. Merging attraction ; A5: Select the merge attraction. Median maximum merger attraction If the value is the largest merging attraction If the value exceeds the attraction threshold, the corresponding node will be assigned to the region. Repeat steps A4 and A5 until no new nodes can be added to the region. Stop the current area. The establishment of the attraction threshold is based on historical aggregated attraction. It has been determined, such as various regions in history. Merge attractiveness when the number of non-core nodes is greater than 2 The average value, if the historical merging attraction If the number is less than the analyzable quantity, an attraction threshold is obtained through manual setting. The analyzable quantity is a manually set value. Whether the attraction threshold is set based on adaptive calculation of historical data or manually, it provides a clear stopping condition for regional expansion, avoids excessive expansion of the region or blurred boundaries, and ensures the independence and stability of each energy storage region. A6: Determine if there are any core nodes that are not marked as seed points; if yes, jump to A3; if no, proceed to the constructed region. Mark them as independent energy storage areas and include those not planned for the area The energy storage devices are marked as independent energy storage areas; The calculation formula (5) is: ; In the formula, For the non-core nodes and regions Each node The sum of their synergistic values area The current number of nodes.

[0028] It is worth noting that the method for determining energy storage regions based on collaborative value in this invention constructs a precise and adaptive dynamic network clustering algorithm. This algorithm can intelligently organize complex energy storage systems into functionally synergistic and clearly defined energy storage regions, laying a solid foundation for subsequent prediction and coordinated control. This method first constructs virtual edges using a collaborative value threshold, forming a weighted network. This step filters out a large number of meaningless or weakly correlated connections, effectively reducing network noise interference and ensuring that subsequent analysis focuses on energy storage device combinations with genuine collaborative potential, thereby improving the purity and purposefulness of region division. Based on this, by introducing the concept of core nodes and identifying them according to their number of connections, it can automatically discover local dense centers in the network. These core nodes naturally constitute the skeleton of potential regions. Compared to traditional random seeds or global center points, this method is more adaptable to the network's own topology, finding more natural starting points for region growth. Subsequently, an iterative region growth strategy based on "merging attraction" is adopted. This attraction index not only considers the sum of the collaborative value between candidate nodes and all existing nodes in the region, but also performs averaging by dividing by the current number of nodes in the region. This makes the merging decision more scientific, assessing not only the correlation strength between nodes and regions, but also preventing the introduction of irrelevant nodes that may be introduced due to the expansion of region size, ensuring the average collaborative quality of members within the region, and maintaining the compactness and cohesion of the region. Finally, through iterative iteration, this algorithm can discover and construct all potential energy storage regions without omission, and automatically marks isolated or weakly correlated points that are not clustered as independent regions. This approach ensures the synergy of the main regions while fully considering the special characteristics of edge devices, forming a comprehensive and hierarchical regional planning scheme. Overall, this method transforms the complex problem of energy storage device relationships into a clear problem of network community discovery. The process is highly automated, robust to noise, and adaptable to network structure. The divided regions exhibit strong internal coordination and clear boundaries between regions. This not only significantly improves the scientific nature and efficiency of regional planning but also directly empowers subsequent charging and discharging prediction and coordinated scheduling. It represents a key technological breakthrough for achieving refined and intelligent management of regional energy storage.

[0029] It should be noted that the region The construction is a dynamic process. Initially, the core node that is not marked as a seed point and has the largest number of virtual edges is marked as a region. When the seed point is in the region nodes There is only one seed point. As non-core nodes are added, the region... The included nodes More and more, until there are no new nodes that can be added to the region. Stop the current area. The establishment of.

[0030] In one possible implementation of this application embodiment, the above-mentioned S2 can be implemented by the following S201, S202, S203, S204, S205 and S206, which are described in detail below: S201: Based on historical electricity consumption forecasts, predict the electricity consumption of each energy storage area on the day containing the target time, including: The charging time is extracted from the regional database, and the non-charging time of the day is marked as the discharge time. The charging time is obtained by setting the historical electricity price for each time period in the n days before the target time, which is generally from 1:00 to 6:00 on each day, and the charging time of the energy storage devices that can supply power to each other is consistent. The regional database is used to store data. Obtain the electricity consumption of each energy storage area within the dischargeable time n days before the target time from the regional database; integrate the electricity consumption into a power consumption group, obtain the variance of the electricity consumption within the power consumption group, and determine whether the variance is greater than the power variance threshold value; where n is obtained through seasonal settings, and the default value is 30. If so, mark the average value of the electricity consumption in the electricity consumption group as the reference electricity consumption, remove the electricity consumption in the electricity consumption group with the largest absolute value of the difference from the reference electricity consumption, recalculate the variance of the remaining electricity consumption in the electricity consumption group and re-perform the variance judgment until the variance of the electricity consumption group is less than the variance threshold value, and calculate the average value of the remaining electricity consumption in the electricity consumption group to obtain the characteristic electricity consumption corresponding to the day of the target time. If not, the average value of the electricity consumption within the electricity consumption group is calculated to obtain the characteristic electricity consumption corresponding to the day of the target time; wherein, the variance threshold is automatically determined based on the statistical distribution of historical electricity consumption data. For example, the daily variance of the electricity consumption during the discharge time in the past M days is calculated to form a variance sample set, and the upper quartile or 90th percentile value of the sample set is taken as the variance threshold, M>60. Obtain the characteristic electricity consumption and actual electricity consumption corresponding to the dischargeable time for n days prior to the target time. Obtain the ratio of actual electricity consumption to characteristic electricity consumption for each day. Calculate the characteristic ratio by averaging several ratios. Multiply the characteristic ratio by the characteristic electricity consumption corresponding to the day of the target time to obtain the predicted electricity consumption corresponding to the dischargeable time for the day of the target time. Wherein, the ratio of actual electricity consumption to characteristic electricity consumption for each day is the actual electricity consumption for each day divided by the characteristic electricity consumption for the corresponding day.

[0031] It is worth noting that the method for predicting the electricity consumption of each energy storage area on the target day based on historical electricity consumption in this invention improves the scientific rigor and foresight of regional energy storage dispatch decisions by constructing a high-precision prediction model that can dynamically adapt to data fluctuations and is robust. This method first intelligently divides the discharge time window by incorporating electricity price information. This timing strategy focuses the prediction on the critical periods when energy storage devices truly participate in grid regulation and energy release, eliminating interference from non-target behaviors such as charging, and ensuring the high relevance and practicality of the predicted data. Meanwhile, this method designs an outlier filtering and feature extraction mechanism based on dynamic variance judgment. Instead of simply using an arithmetic mean, it iteratively removes data points with the largest deviation from the mean until the variance of the dataset converges below a threshold. This process effectively addresses anomalies such as peak or trough electricity consumption caused by sudden events, special weather, or temporary loads, avoiding excessive distortion of the prediction benchmark by these outliers. This ensures that the extracted "featured electricity consumption" accurately and stably reflects the electricity consumption level benchmark of the region under normal operating conditions, greatly enhancing the robustness of the prediction model to data noise. Furthermore, the variance threshold is not a fixed empirical constant but is adaptively determined based on statistical analysis of massive historical data. This means the model can learn and adapt to the inherent fluctuations in electricity consumption patterns across different seasons and regions, ensuring that the outlier judgment criteria match the actual data distribution characteristics. This avoids the risk of misjudgment caused by a "one-size-fits-all" approach and demonstrates a high degree of adaptability. Finally, this method introduces a "characteristic ratio" for secondary correction. This ratio is calculated by averaging the historical actual electricity consumption with the characteristic electricity consumption of the same day, capturing regular systematic deviations or recent trend changes, such as the gradual expansion of production scale of enterprises in the region and the increase or decrease of air conditioning load due to seasonal temperature changes. This proportional correction factor integrates inherent and stable deviation patterns from history into the final prediction, ensuring that the prediction results are not only based on a cleaned and stable benchmark but also incorporate precise calibration of dynamic trends, thus achieving a leap from "static benchmark" to "dynamic prediction." In summary, this method, through intelligent time period division, robust data cleaning, adaptive threshold setting, and proportional correction, forms a closed-loop, progressive prediction process. The prediction results can effectively filter out random noise, accurately capture core patterns, and dynamically respond to trend changes, providing highly reliable decision input for subsequent integrated energy storage modules. This is a key technological innovation for improving the economic benefits of regional energy storage and the grid's support capabilities.

[0032] It should be noted that, among the electricity consumption groups with the largest absolute difference from the reference electricity consumption, if the electricity consumption group with the largest absolute difference from the reference electricity consumption also has the largest and smallest electricity consumption, then the largest electricity consumption will be removed first.

[0033] It should be noted that if, after removing 90% of the electricity consumption, the variance of the remaining electricity consumption is still not less than the electricity variance threshold, then the average value of the remaining 10% of the electricity consumption group is used as the characteristic electricity consumption.

[0034] S202: Based on predicted electricity consumption and maximum storage capacity, each energy storage area is divided into a first zone, a second zone, and a third zone, including: The maximum stored power is obtained by adding up the maximum stored power of each energy storage device in the energy storage area, and the predicted power consumption corresponding to each energy storage area is extracted. When the maximum stored power of the energy storage area is greater than the predicted power consumption, the energy storage area is marked as the first area; when the maximum stored power of the energy storage area is less than the predicted power consumption, the energy storage area is marked as the second area; when the maximum stored power of the energy storage area is equal to the predicted power consumption, the energy storage area is marked as the third area.

[0035] S203: Identify the target second region of the first region, including: B1: Subtract the predicted power consumption from the maximum stored power in the first region to obtain the surplus power in the first region. The predicted power consumption of the second region is subtracted from the maximum stored power to obtain the power deficit of the second region. ;in, This is the number of the first region. The numbering of the second area; B2: Based on surplus power and low battery The power supply compatibility of each first and second region is determined by formula (6). In this step, when there is surplus power = Low battery hour, For a perfect match, if there is surplus power... With low battery The differences are huge, then Approaching 0; B3: Extract the number from the regional database. The first area and numbered Power loss rate between the second region Based on power loss rate and power supply compatibility The number is determined by formula (7). The first area and numbered Power supply potential index between the second region Among them, the power loss rate is based on the numbered The first area and numbered The line length and line impedance between the second region are calculated, and the line length is obtained from the numbered region. The center of the first area is numbered as follows The length of the line between the centers of the second region; B4: Based on surplus power Sort the first region in descending order of quantity; B5: Extract the first region sequentially based on the sorting number from smallest to largest, and then extract the power supply potential index corresponding to the first region. The index numbers are obtained by sorting them from largest to smallest. The index numbers are then updated after removing the index numbers corresponding to the target second region that have been marked as other first regions. B6: Extract the index numbers in ascending order based on the updated index numbers, and determine whether the current index number is greater than the index number threshold. If so, mark the second region corresponding to the index number before the index number threshold as the target second region of the current first region, and jump to B5 to traverse the remaining first regions. The index number threshold is a threshold used to avoid the number of second regions that are over-supplied by a first region. It is a parameter that controls the matching complexity and communication overhead. The value can be set according to the scale of energy storage devices in the first region, and is generally set to 6. If not, compare the current index number with the corresponding under-charge of the index numbers preceding the current index number. The total deficit is obtained by dividing by the corresponding power loss difference rate and then adding them together; when the total deficit is less than the current surplus power of the first region... Repeat step B6; when the total under-charge is not less than the current surplus charge in the first region. At that time, mark the second region corresponding to the index number before the current index number as the target second region of the current first region, and jump to B5 to traverse the remaining first regions; where the power loss difference rate is one minus the power deficit. The power loss rate corresponding to the second region and the current first region The value of, i.e. The value; The calculation formula (6) is: ; The calculation formula (7) is: ; In the formula, The baseline loss rate is used to convert different units of measurement. After normalization, the power loss rate between each first region and the second region can be obtained. The average value; and It is the proportional adjustment coefficient, and It can be dynamically adjusted according to the system operation strategy. For example, in a supply guarantee scenario, the capacity can be appropriately increased. Prioritize ensuring sufficient power supply; when pursuing economic benefits, the capacity can be appropriately increased. Prioritize low-loss paths.

[0036] It is worth noting that this invention, by identifying the target second region within the first region, constructs a multi-dimensional intelligent matching decision-making system. This system can accurately and quickly select the optimal power supply combination from numerous potential supply and demand relationships, thereby maximizing the benefits of energy mutual assistance between regions. This method first quantifies the core concept of "power supply adaptability." By measuring the degree of fit between the surplus power of the supplier and the insufficient power of the demander, the abstract supply and demand matching problem is transformed into a calculable optimization index, making it possible to find the "perfect" supply and demand pair and laying a solid quantitative foundation for subsequent refined screening. This method further introduces a "power supply potential index," which not only integrates the power supply adaptability representing the accuracy of supply and demand matching but also creatively couples the power loss rate representing the economic efficiency and physical feasibility of energy transmission. The two are unified through weighted index calculation, particularly through dynamically adjustable weighting coefficients. and System administrators can adjust their strategies based on real-time operational plans. For example, during periods of high load, they can prioritize increasing the weight of power supply adaptability to ensure grid security, or during normal operation periods, they can focus on the weight of loss rate to pursue economic benefits. This achieves a flexible shift in decision-making objectives from a single dimension to a multi-dimensional trade-off, enabling energy dispatch strategies to respond flexibly to complex and ever-changing grid demands. In terms of the specific matching algorithm, this method employs an efficient "greedy" sorting and iterative allocation strategy. First, it sorts suppliers by surplus power, prioritizing areas with the most abundant energy, ensuring that major energy suppliers receive the best allocation opportunities and improving overall dispatch efficiency. Then, it sorts the potential demanders of each supplier by power supply potential index, and combines this with a sequence threshold and a dynamic total under-supply verification mechanism. This not only effectively controls the matching complexity and communication overhead of individual suppliers through the sequence threshold, avoiding network congestion and management chaos, but also ensures that the output capacity of suppliers is fully utilized through real-time verification of accumulated under-supply, avoiding resource idleness or over-commitment. In particular, the introduction of power loss difference rate correction when calculating the total power deficit extends the line loss factor from the selection criteria to the allocation strategy, ensuring that the final allocated power is an effective supply after considering actual transmission losses, thus guaranteeing the closed-loop and accuracy of the decision-making process. Ultimately, this method, through global traversal and exclusive labeling, forms a conflict-free and highly efficient global optimization scheme. It not only maximizes the absorption of regional surplus power and accurately fills the power gap, but also reduces overall losses and controls scheduling costs at the system level, providing crucial technical support for building an efficient, collaborative, and reliable regional energy internet.

[0037] It should be noted that the current index number corresponds to the same amount of electricity deficit as the index numbers preceding it. The total shortfall is obtained by dividing by the corresponding power loss difference rate and then summing them up. Specifically, it is the shortfall with an index number. Divide by the corresponding power loss difference rate to get a loss-calculated under-power amount, and add up several loss-calculated under-power amounts to get the total under-power amount.

[0038] It should be noted that in steps B5 and B6, if there is a case where the index number corresponding to the first region is 0, then the marking of the target second region of the current first region is stopped.

[0039] S204: Determine the power adjustment instruction for the corresponding first region based on the insufficient power in the target second region, including: Extract the target second region from the first region, and determine the low power of the target second region. The total deficit power of the target second region is obtained by dividing by the corresponding power loss difference rate and then adding them together; when the total deficit power is equal to the surplus power... When the ratio exceeds the ratio threshold one, power adjustment command one is issued; among them, ratio threshold one, ratio threshold two and ratio threshold three are key parameters for classifying the system response level, and are ordered from largest to smallest as ratio threshold one, ratio threshold two and ratio threshold three. Their values ​​can be optimized according to system operation strategy and historical data. For example, the impact of different threshold settings on the total system revenue or power supply reliability under different historical supply and demand scenarios can be analyzed by simulation, and the threshold combination that makes the overall benefit optimal can be selected. When the total short power and the surplus power When the ratio exceeds the ratio threshold 2 but does not exceed the ratio threshold 1, power adjustment command 2 is issued; When the total short power and the surplus power When the ratio does not exceed the ratio threshold two, power adjustment command three is issued.

[0040] S205: Determine the real-time charging power of the energy storage device in the first area based on the power adjustment command, including: When the power adjustment command is power adjustment command one, the energy storage device in the first area is charged at maximum power, and charging stops when the power stored in the energy storage device in the first area is greater than the sum of the current predicted power consumption in the first area and the total power deficit in the corresponding target second area. When the power adjustment command is power adjustment command two, the energy storage devices in the first region are charged at maximum power before the reference time point within the charging time, and at adaptive power after the reference time point within the charging time. Charging stops when the stored power in the energy storage devices in the first region exceeds the sum of the predicted power consumption of the first region and the total shortfall in the corresponding target second region. The reference time point is manually set and is generally located at 70% of the charging time. The specific calculation method for adaptive power charging is as follows: the sum of the predicted power consumption of the first region and the total shortfall in the corresponding target second region is marked as the required power; the required power is subtracted from the total existing power of the energy storage devices in the first region to obtain the region's uncharged amount; the storable power of each energy storage device in the first region is divided by the total storable power of the first region to obtain the power allocation ratio; the region's uncharged amount is multiplied by the allocation ratio to obtain the uncharged amount of each energy storage device; the duration between the current time and the end time of the charging time is marked as the charging duration; the uncharged amount is divided by the charging duration to obtain the adaptive power charging for each energy storage device. When the power adjustment command is power adjustment command three, the energy storage device in the first area is charged with adaptive power, and charging stops when the power stored in the energy storage device in the first area is greater than the sum of the current predicted power consumption of the first area and the total power deficit of the corresponding target second area.

[0041] It should be noted that, in determining the real-time charging power of the energy storage device in the first area based on the power adjustment command, the charging power corresponding to adjustment command one, adjustment command two and adjustment command three can all be set to the maximum charging power.

[0042] S206: Discharge the corresponding first region based on the real-time stored energy in the second target region, including: The target second regions in the first region are sorted from largest to smallest amount of electricity deficit, and based on the sorting, electricity equal to the amount of electricity deficit is supplied to the target second regions through the first region in sequence.

[0043] It should be noted that the equivalent shortfall is the amount of electricity equal to the shortfall in the target second area.

[0044] It should be noted that discharging the corresponding first region based on the real-time stored power of the second target region can also be: when the real-time stored power of the second target region is less than 0, discharging to the current second target region through the corresponding first region.

[0045] The above primarily describes the solutions of the embodiments of this application from the perspective of device implementation. It is understood that each device, for example, a charge / discharge prediction system for regional energy storage, includes at least one of the hardware structures and software modules corresponding to the execution of each function in order to achieve the above-mentioned functions. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0046] This application embodiment can divide a charge-discharge prediction system for regional energy storage into functional units based on the above method example. For example, each function can be divided into separate functional units, or two or more functions can be integrated into a single linked energy storage module. The integrated units can be implemented in hardware or as software functional units. It should be noted that the unit division in this application embodiment is illustrative and only represents a logical functional division; other division methods may be used in actual implementation.

[0047] When using integrated units, Figure 3 A possible structural schematic diagram of a charge and discharge prediction system (referred to as energy storage management system 30) for regional energy storage involved in the above embodiments is shown. The energy storage management system 30 includes a linkage energy storage module 301 and a regional planning module 302, and may also include a storage medium (referred to as storage unit 303). Figure 3 The schematic diagram shown can be used to illustrate the structure of a charge and discharge prediction system for regional energy storage involved in the above embodiments.

[0048] when Figure 3 The schematic diagram shown illustrates the structure of a charge-discharge prediction system for regional energy storage involved in the above embodiments. The linkage energy storage module 301 is used to control and manage the operation of the charge-discharge prediction system for regional energy storage. The regional planning module 302 is used for the charge-discharge prediction system for regional energy storage to communicate with other devices. The storage unit 303 is used to store the program code and data of the charge-discharge prediction system for regional energy storage.

[0049] For example, the regional planning module 302 is used to obtain the released power and similar received power of each energy storage device the day before the target time, and to plan the energy storage devices based on the released power and similar received power to obtain the energy storage area; the similar received power is the power received by the power supply area corresponding to the energy storage device from other areas of energy storage. Linked energy storage module 301: It is used to predict the predicted electricity consumption of each energy storage area on the day of the target time based on historical electricity consumption, divide each energy storage area into a first area, a second area and a third area based on the predicted electricity consumption and the maximum stored capacity; identify the target second area of ​​the first area, determine the corresponding power adjustment command of the first area based on the insufficient power of the target second area, determine the real-time charging power of the energy storage device in the first area based on the power adjustment command, and discharge the corresponding first area based on the real-time stored capacity of the target second area.

[0050] Although this application has been described herein in conjunction with various embodiments, those skilled in the art, by reviewing the accompanying drawings, disclosure, and appended claims, will understand and implement other variations of the disclosed embodiments in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple instances. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0051] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely illustrative descriptions of the application as defined by the appended claims, and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from the spirit and scope of this application. Thus, if such modifications and modifications fall within the scope of the claims of this application and their equivalents, this application is also intended to include such modifications and modifications.

[0052] Some of the data in the above calculation formula are obtained by removing dimensions and taking their numerical values. The calculation formula is a calculation formula that is closest to the real situation, obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the calculation formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

Claims

1. A method for predicting charge and discharge for regional energy storage, characterized in that, include: S1: Obtain the released power and received power of each energy storage device the day before the target time, and plan the energy storage devices based on the released power and received power to obtain the energy storage area; where the received power is the power received by the power supply area corresponding to the energy storage device from other areas. S2: Based on historical electricity consumption, predict the predicted electricity consumption of each energy storage area on the day of the target time. Divide each energy storage area into a first area, a second area, and a third area based on the predicted electricity consumption and the maximum stored capacity. Identify the target second area of ​​the first area. Determine the corresponding power adjustment command for the first area based on the insufficient power of the target second area. Determine the real-time charging power of the energy storage device in the first area based on the power adjustment command. Discharge the corresponding first area based on the real-time stored capacity of the target second area.

2. The method for predicting charge and discharge for regional energy storage according to claim 1, characterized in that, The energy storage area is obtained by planning the energy storage device based on the released electricity and the similar received electricity, including: Release of electricity from each energy storage device and class receive power The standard value is obtained by calculating formula (1). and standard value Based on standard values and standard value The stability of net power outflow is determined by formula (2). Through standard values Standard value and stability of net power outflow Constructing 3D feature vectors ;in, The serial number of the energy storage device; Based on the three-dimensional feature vector corresponding to the energy storage device The number is determined by formula (3). The energy storage device and its number are Behavioral similarity between energy storage devices Among them, the storage number is The three-dimensional feature vectors corresponding to the energy storage devices are respectively Number The three-dimensional feature vector corresponding to the energy storage device is , The serial number of the energy storage device; Obtain the latitude and longitude of each energy storage device, and determine its number based on the latitude and longitude using formula (4). The energy storage device and its number are Geographic proximity between energy storage devices ; behavioral similarity Geographic proximity The synergistic value is obtained by adding the weights together. The weights are added together as follows: , It is a proportional adjustment coefficient determined based on the manager's preference for behavioral roles and geographical proximity, and The value range is (0.1, 0.9); Based on collaborative value Identify energy storage areas; The calculation formula (1) is: ； In the formula, To release electricity Standardized value For receiving electricity The standardized value; The maximum value of the released electricity from all energy storage devices. This represents the minimum amount of electricity released by all energy storage devices. The maximum value of the received electricity among all energy storage devices. This is the minimum value of the received electricity among all energy storage devices. The calculation formula (2) is: ； In the formula, To prevent extremely small positive numbers from being divided by zero; The calculation formula (3) is: ； The calculation formula (4) is: ； In the formula, For the number The energy storage device and its number are The Euclidean distance between energy storage devices is determined by latitude and longitude. It is a scale parameter used to control the rate at which proximity decays with distance, and is determined based on the geographical distance between all energy storage devices.

3. The method for predicting charge and discharge for regional energy storage according to claim 2, characterized in that, The based on collaborative value Determine the energy storage area, including: A1: Treating each energy storage device as a node, if the collaborative value between two nodes... If the value is greater than the collaborative value threshold, a virtual edge is established between the two nodes; wherein the weight of the virtual edge is the corresponding collaborative value. The value; the synergy value threshold is determined by analyzing all synergy values. Choose a collaborative value that can filter out the bottom 20% or 30% of the values. As a threshold for collaborative value; A2: Traverse all nodes and mark nodes with more than a certain number of virtual edges as core nodes; where the number threshold is set by the number of energy storage devices. A3: Mark the core node that is not marked as a seed point and has the largest number of virtual edges as a region. Seed points; among them, It is the area code; A4: Find all regions any internal node Non-core nodes that are connected by virtual edges and have not yet been classified are determined by formula (5) to identify several of the aforementioned non-core nodes and regions. Merging attraction ; A5: Select the merge attraction. Median maximum merger attraction If the value is the largest merging attraction If the value exceeds the attraction threshold, the corresponding node will be assigned to the region. Repeat steps A4 and A5 until no new nodes can be added to the region. Stop the current area. The establishment of the attraction threshold is based on historical aggregated attraction. It is confirmed; A6: Determine if there are any core nodes that are not marked as seed points; if yes, jump to A3; if no, proceed to the constructed region. Mark them as independent energy storage areas and include those not planned for the area The energy storage devices are marked as independent energy storage areas; The calculation formula (5) is: ； In the formula, For the non-core nodes and regions Each node The sum of their synergistic values area The current number of nodes.

4. The method for predicting charge and discharge for regional energy storage according to claim 1, characterized in that, The historical electricity consumption forecast for each energy storage area on the target day includes: The charging time is extracted from the regional database, and the non-charging time of the day is marked as the discharge time; the charging time is obtained by setting the historical electricity price for each time period in the n days prior to the target time. Obtain the electricity consumption of each energy storage area within the dischargeable time n days before the target time from the regional database; integrate the electricity consumption into a power consumption group, obtain the variance of the electricity consumption within the power consumption group, and determine whether the variance is greater than the power variance threshold value. If so, mark the average value of the electricity consumption in the electricity consumption group as the reference electricity consumption, remove the electricity consumption in the electricity consumption group with the largest absolute value of the difference from the reference electricity consumption, recalculate the variance of the remaining electricity consumption in the electricity consumption group and re-perform the variance judgment until the variance of the electricity consumption group is less than the variance threshold value, and calculate the average value of the remaining electricity consumption in the electricity consumption group to obtain the characteristic electricity consumption corresponding to the day of the target time. If not, the characteristic electricity consumption corresponding to the day of the target time is obtained by averaging the electricity consumption within the electricity consumption group; wherein, the variance threshold is automatically determined based on the statistical distribution of historical electricity consumption data. Obtain the characteristic electricity consumption and actual electricity consumption corresponding to the dischargeable time for n days prior to the target time. Obtain the ratio of actual electricity consumption to characteristic electricity consumption for each day. Calculate the characteristic ratio by averaging several ratios. Multiply the characteristic ratio by the characteristic electricity consumption corresponding to the day of the target time to obtain the predicted electricity consumption corresponding to the dischargeable time for the day of the target time. Wherein, the ratio of actual electricity consumption to characteristic electricity consumption for each day is the actual electricity consumption for each day divided by the characteristic electricity consumption for the corresponding day.

5. The method for predicting charge and discharge for regional energy storage according to claim 1, characterized in that, The division of each energy storage area into a first area, a second area, and a third area based on predicted electricity consumption and maximum storage capacity includes: The maximum stored power is obtained by adding up the maximum stored power of each energy storage device in the energy storage area, and the predicted power consumption corresponding to each energy storage area is extracted. When the maximum stored power of the energy storage area is greater than the predicted power consumption, the energy storage area is marked as the first area; when the maximum stored power of the energy storage area is less than the predicted power consumption, the energy storage area is marked as the second area; when the maximum stored power of the energy storage area is equal to the predicted power consumption, the energy storage area is marked as the third area.

6. The method for predicting charge and discharge for regional energy storage according to claim 1, characterized in that, The target second region for identifying the first region includes: B1: Subtract the predicted power consumption from the maximum stored power in the first region to obtain the surplus power in the first region. The predicted power consumption of the second region is subtracted from the maximum stored power to obtain the power deficit of the second region. ;in, This is the number of the first region. The numbering of the second area; B2: Based on surplus power and low battery The power supply compatibility of each first and second region is determined by formula (6). ; B3: Extract the number from the regional database. The first area and numbered Power loss rate between the second region Based on power loss rate and power supply compatibility The number is determined by calculation formula (7). The first area and numbered Power supply potential index between the second region Among them, the power loss rate is based on the numbered The first area and numbered The line length and line impedance between the second region are calculated. B4: Based on surplus power Sort the first region in descending order of quantity; B5: Extract the first region sequentially based on the sorting number from smallest to largest, and then extract the power supply potential index corresponding to the first region. The index numbers are obtained by sorting them from largest to smallest. The index numbers are then updated after removing the index numbers corresponding to the target second region that have been marked as other first regions. B6: Extract the index numbers in ascending order based on the updated index number, and determine whether the current index number is greater than the index number threshold; if so, mark the second region corresponding to the index number before the index number threshold as the target second region of the current first region, and jump to B5 to traverse the remaining first regions; where the index number threshold is a threshold for the number of second regions to avoid a first region supplying too much power. If not, compare the current index number with the corresponding under-charge of the index numbers preceding the current index number. The total deficit is obtained by dividing by the corresponding power loss difference rate and then adding them together; when the total deficit is less than the current surplus power of the first region... Repeat step B6; when the total under-charge is not less than the current surplus charge in the first region. At that time, mark the second region corresponding to the index number before the current index number as the target second region of the current first region, and jump to B5 to traverse the remaining first regions; where the power loss difference rate is one minus the power deficit. The power loss rate corresponding to the second region and the current first region The value; The calculation formula (6) is: ； The calculation formula (7) is: ； In the formula, As the baseline loss rate, and It is the proportional adjustment coefficient, and .

7. The method for predicting charge and discharge for regional energy storage according to claim 6, characterized in that, The determination of the power adjustment instruction for the corresponding first region based on the insufficient power in the target second region includes: Extract the target second region from the first region, and determine the low power of the target second region. The total deficit power of the target second region is obtained by dividing by the corresponding power loss difference rate and then adding them together; when the total deficit power is equal to the surplus power... When the ratio exceeds the ratio threshold, a power adjustment command is issued. When the total short power and the surplus power When the ratio exceeds the ratio threshold 2 but does not exceed the ratio threshold 1, power adjustment command 2 is issued; When the total short power and the surplus power When the ratio does not exceed the ratio threshold two, power adjustment command three is issued.

8. The method for predicting charge and discharge for regional energy storage according to claim 7, characterized in that, The determination of the real-time charging power of the energy storage device in the first area based on the power adjustment command includes: When the power adjustment command is power adjustment command one, the energy storage device in the first area is charged at maximum power, and charging stops when the power stored in the energy storage device in the first area is greater than the sum of the current predicted power consumption in the first area and the total power deficit in the corresponding target second area. When the power adjustment command is power adjustment command two, the energy storage devices in the first region are charged at maximum power before the reference time point within the charging time, and at adaptive power after the reference time point within the charging time. Charging stops when the stored power in the energy storage devices in the first region exceeds the sum of the predicted power consumption of the first region and the total shortfall in the corresponding target second region. The specific calculation method for adaptive power charging is as follows: the sum of the predicted power consumption of the first region and the total shortfall in the corresponding target second region is marked as the required power; the required power is subtracted from the total existing power of the energy storage devices in the first region to obtain the region's uncharged amount; the storable power of each energy storage device in the first region is divided by the total storable power of the first region to obtain the power allocation ratio; the region's uncharged amount is multiplied by the allocation ratio to obtain the uncharged amount of each energy storage device; the duration between the current time and the end time of the charging time is marked as the rechargeable duration; the uncharged amount is divided by the rechargeable duration to obtain the adaptive power charging for each energy storage device. When the power adjustment command is power adjustment command three, the energy storage device in the first area is charged with adaptive power, and charging stops when the power stored in the energy storage device in the first area is greater than the sum of the current predicted power consumption of the first area and the total power deficit of the corresponding target second area.

9. The method for predicting charge and discharge for regional energy storage according to claim 1, characterized in that, The step of discharging the corresponding first region based on the real-time stored power in the second target region includes: The target second regions in the first region are sorted from largest to smallest amount of electricity deficit, and based on the sorting, electricity equal to the amount of electricity deficit is supplied to the target second regions through the first region in sequence.

10. A charge / discharge prediction system for regional energy storage, used to operate and implement the charge / discharge prediction method for regional energy storage according to any one of claims 1 to 9, characterized in that, Includes: regional planning module and integrated energy storage module; The regional planning module is used to obtain the released power and similar received power of each energy storage device the day before the target time, and to plan the energy storage devices based on the released power and similar received power to obtain the energy storage area; the similar received power is the power received by the power supply area corresponding to the energy storage device from other areas of energy storage. The linked energy storage module is used to predict the predicted electricity consumption of each energy storage area on the day of the target time based on historical electricity consumption, divide each energy storage area into a first area, a second area, and a third area based on the predicted electricity consumption and the maximum stored capacity; identify the target second area of ​​the first area, determine the corresponding power adjustment command for the first area based on the insufficient power of the target second area, determine the real-time charging power of the energy storage device in the first area based on the power adjustment command, and discharge the corresponding first area based on the real-time stored capacity of the target second area.

Images