Real-time collaborative regulation system and method for large-scale user-side distributed energy storage cluster

Abstract

The application discloses a large-scale user-side distributed energy storage cluster real-time collaborative regulation system and method, and belongs to the technical field of smart grids. The system comprises a distributed energy storage dynamic grouping module, a dispatching instruction generation module, a global collaborative regulation target generation module, a refined power regulation module and a target collaborative instruction output module. The distributed energy storage devices are dynamically grouped to obtain energy storage subgroups. The total power demand is evaluated for consistency to determine the grid dispatching instruction. The global collaborative regulation target is generated according to the grid dispatching instruction and the local voltage-frequency overrun event of the energy storage subgroups. The edge collaborative nodes are distributedly optimized to obtain the refined power adjustment instruction of the distributed energy storage devices. The current real-time operation state is checked for safety to obtain the target collaborative instruction of the large-scale user-side energy storage cluster. The application can realize the real-time collaborative regulation of the user-side energy storage cluster, and improve the regulation accuracy, response efficiency and equipment operation safety.

Description

Technical Field

[0001] This invention belongs to the field of smart grid technology, and in particular relates to a real-time collaborative control system and method for large-scale user-side distributed energy storage clusters. Background Technology

[0002] With the advancement of dual-carbon goals and the deepening construction of new power systems, user-side distributed energy storage devices are showing a trend towards large-scale and decentralized deployment. Massive numbers of energy storage units are connected to the distribution network, providing considerable flexibility resources for power regulation, peak shaving and valley filling, and voltage and frequency support. To leverage the aggregated value of distributed energy storage devices, existing control systems typically employ a pre-defined static grouping approach for unified management. Power regulation commands are issued by the upper-level dispatch platform, and each group distributes regulation tasks according to fixed rules. Some systems also introduce edge nodes for local control of devices within a group to alleviate the communication and computing pressure of centralized control.

[0003] However, the aforementioned static and extensive control methods are ill-suited to the real-time control requirements of large-scale energy storage clusters, exhibiting the following problems: First, equipment grouping is not dynamically adjusted based on real-time operating characteristics such as state of charge and voltage levels, failing to effectively exploit the aggregated response characteristics of energy storage devices and resulting in a lack of targeted power regulation allocation; Second, the generation of grid dispatch commands does not systematically encode the response speed and duration characteristics of the equipment, leading to low command transmission and execution efficiency; Third, the lack of a globally coordinated control target generation mechanism means that grid dispatch requirements are not effectively linked to local voltage-frequency exceedance events, easily causing conflicts between power regulation and electrical anomaly handling; Fourth, the refined control process lacks a distributed collaborative optimization process with iterative interaction among devices, making it difficult for regulation components to achieve coordinated convergence in power matching and response consistency; Fifth, safety verification only considers a few operating parameters, failing to integrate multi-dimensional information such as maximum chargeable / dischargeable power, state of charge, and operating temperature, easily generating control commands that exceed the safe operating range of the equipment, reducing control accuracy while increasing equipment operating risks. Therefore, how to achieve real-time collaborative control of large-scale user-side distributed energy storage clusters has become an urgent technical problem to be solved. Summary of the Invention

[0004] In view of the shortcomings of the prior art, the purpose of this invention is to provide a real-time collaborative control system and method for large-scale user-side distributed energy storage clusters. Through dynamic clustering, scheduling instruction encoding, global collaborative target generation, distributed collaborative optimization and multi-dimensional security verification, the system achieves real-time collaborative control of user-side energy storage clusters, thereby improving control accuracy, response efficiency and equipment operation safety.

[0005] To achieve the above objectives, the present invention provides a large-scale user-side distributed energy storage cluster real-time collaborative control system, comprising the following modules connected in sequence: The distributed energy storage dynamic clustering module is used to dynamically cluster distributed energy storage devices based on real-time operating status data and grid connection point electrical quantity data, thereby obtaining energy storage subgroups of distributed energy storage devices. The dispatch instruction generation module is used to perform a consistency assessment of the total power demand of the energy storage subgroup based on the grid dispatch requirements and the division results of the energy storage subgroup, and determine the grid dispatch instructions for the energy storage subgroup based on the assessment results and the equipment parameters of the energy storage subgroup. The global coordinated control target generation module is used to generate global coordinated control targets for the energy storage subgroup based on grid dispatch instructions and local voltage-frequency over-limit events of the energy storage subgroup. The refined power regulation module is used to perform distributed collaborative optimization on the edge collaborative nodes of the energy storage subgroup based on the global collaborative regulation target, so as to obtain the refined power adjustment command of the distributed energy storage device. The target coordination instruction output module is used to perform a safety verification between the fine-grained power adjustment instruction and the current real-time operating status of the distributed energy storage device based on the maximum chargeable and dischargeable power of the distributed energy storage device, so as to obtain the target coordination instruction of the large-scale user-side energy storage cluster.

[0006] Preferably, in the distributed energy storage dynamic clustering module, the process of obtaining the energy storage subgroup of the distributed energy storage devices is as follows: Based on real-time operating status data and grid connection point electrical quantity data, determine the state of charge range and voltage level of the distributed energy storage equipment; Based on the state of charge range and voltage level, a correlation analysis is performed on the current state of charge and available power capacity of distributed energy storage devices to construct a voltage sensitivity matrix for distributed energy storage devices. Clustering iteration is performed on the eigenvectors in the voltage sensitivity matrix, and distributed energy storage devices whose eigenvectors are close in distance in the feature space are grouped into the same set of devices to obtain the initial set of distributed energy storage devices. The current state of charge and available power capacity of each device in the initial device set are statistically evaluated to obtain the aggregate response characteristic parameters of the initial device set; Assign a unique virtual energy storage identifier to the initial set of devices with aggregated response characteristic parameters to obtain the energy storage subgroup of distributed energy storage devices.

[0007] Preferably, in the dispatch instruction generation module, the process of determining the grid dispatch instruction for the energy storage subgroup is as follows: Obtain the power grid dispatching requirements issued by the superior dispatching platform. The power grid dispatching requirements include the total regional power regulation within the target time window. Based on the division results of the energy storage subgroups, the available power capacity of the distributed energy storage devices in the energy storage subgroups is statistically analyzed to obtain the adjustable power range of the energy storage subgroups. The total regional power regulation is matched with the adjustable power range of the energy storage subgroups to determine the power regulation share that each energy storage subgroup needs to undertake, and the sum of the power regulation shares of the energy storage subgroups is kept consistent with the total regional power regulation. Extract the response speed level and response duration characteristics of each energy storage subgroup from the equipment parameters of the energy storage subgroup; Power correlation encoding is performed on the characteristics of power regulation share, response speed level, and response duration to obtain the grid dispatch instructions for the energy storage subgroup. The grid dispatch instructions include the power regulation direction and power regulation amount that the energy storage subgroup needs to perform.

[0008] Preferably, in the scheduling instruction generation module, the power regulation share, response speed level, and response duration characteristics are encoded using power correlation as follows: A synergistic analysis of the power injection and power consumption demands of the energy storage subgroup within the power regulation share is conducted to determine the power regulation direction of the energy storage subgroup. Based on the amplitude characteristics of the power regulation share, the power regulation demand of the energy storage subgroup is converted accordingly to obtain the power regulation level of the power regulation share. Based on the response speed level, the power regulation response speed attribute in the energy storage subgroup is labeled and analyzed to obtain the response mode identifier of the energy storage subgroup. Based on the response duration characteristics, the power support period level of the energy storage subgroup is determined. The power support period level is used to characterize the duration for which the energy storage subgroup can maintain power output. By associating and combining the power regulation direction, power regulation level, response mode identifier, and power support time period, the grid dispatch instructions for the energy storage subgroup are obtained.

[0009] Preferably, in the scheduling instruction generation module, the process of obtaining the power regulation level of the power regulation share is as follows: Extract the power regulation share value of each energy storage subgroup from the power regulation share, and extract the maximum adjustable power capacity of each energy storage subgroup from the equipment parameters of the energy storage subgroup. Obtain the total number of pre-set power regulation levels, which is determined based on the regulation accuracy requirements of the energy storage cluster; Based on the power regulation share, maximum adjustable power capacity, and total number of levels, the power regulation level of the energy storage subgroup is calculated using the following formula: ; The power regulation level for the i-th energy storage subgroup, with a value range of... , Let i be the power regulation share value of the i-th energy storage subgroup. Let N be the maximum adjustable power capacity of the i-th energy storage subgroup, and N be the total number of preset power regulation levels. This is for floor function.

[0010] Preferably, in the global coordinated regulation target generation module, the process of generating the global coordinated regulation target for the energy storage subgroup is as follows: The over-limit type, over-limit magnitude, and over-limit duration of each energy storage subgroup are aggregated into local voltage-frequency over-limit event information for the energy storage subgroup; The power grid dispatching command is decoded and separated to obtain the power regulation requirements of the power grid dispatching command. The power regulation requirements include the power regulation direction, power regulation amount and regulation response time period. The power regulation requirements are correlated and compared with local voltage-frequency over-limit event information to obtain the regulation-event correlation degree identifier of the energy storage subgroup; Based on the regulation-event correlation identifier, the power regulation requirements of each subgroup in the energy storage subgroup are prioritized to obtain the regulation priority sequence of power regulation requirements; The power regulation requirements of different energy storage subgroups in the regulation priority sequence are matched and merged to obtain the global coordinated regulation target of the energy storage subgroups. The global coordinated regulation target includes the total power regulation amount and the overall response time window.

[0011] Preferably, in the refined power control module, the process of obtaining the refined power adjustment command for the distributed energy storage device is as follows: It receives global coordinated control targets and obtains the current real-time operating status of each distributed energy storage device in the energy storage subgroup; Based on the edge collaborative nodes of the energy storage subgroup, combined with the global collaborative control target and the current real-time operating status, the initial power regulation requirement of the energy storage subgroup is obtained, and the initial power regulation requirement is decomposed into preliminary regulation components corresponding to each distributed energy storage device. The initial adjustment components are sent to the corresponding distributed energy storage devices, and the acceptance feedback information returned by the distributed energy storage devices through their own real-time constraints is received. Based on the feedback information, the initial adjustment components are iteratively and interactively adjusted until the adjustment components of the distributed energy storage device reach the preset collaborative convergence state in terms of power matching degree and response consistency. Each adjustment component after reaching the collaborative convergence state is determined as a fine-grained power adjustment command for each distributed energy storage device, and the fine-grained power adjustment command is sent to the corresponding distributed energy storage device.

[0012] Preferably, in the refined power control module, the initial adjustment component is iteratively and interactively adjusted until the adjustment component of the distributed energy storage device reaches a preset cooperative convergence state in terms of power matching degree and response consistency. Specifically: Read the aggregated response characteristic parameters of the distributed energy storage device and the cluster collaborative control requirements of the edge collaborative nodes respectively; The edge collaboration node interacts and adjusts with each distributed energy storage device in the energy storage subgroup to obtain the current state of charge, adjustability and local electrical safety boundary information of the distributed energy storage device; Based on the requirements for cluster coordinated control and combined with the aggregation response characteristics of the energy storage subgroup, the power sharing intentions of the equipment in the energy storage subgroup are initially proposed, and the initial power sharing intention set of the energy storage subgroup is obtained. The initial power sharing intention set is exchanged with the edge collaborative nodes of the adjacent energy storage subgroups, and the power sharing intention information of the adjacent energy storage subgroups is received. Based on the power sharing intention information, load judgment is made on the local network overload and equipment charge status of large-scale user-side energy storage clusters, and the power sharing intention is adaptively adjusted according to the judgment results.

[0013] Preferably, in the target coordination instruction output module, the process of obtaining the target coordination instruction for the large-scale user-side energy storage cluster is as follows: The maximum chargeable and dischargeable power is extracted from the equipment parameters of the distributed energy storage device to obtain the power limit set of the distributed energy storage device. The power limit set includes the upper limit value of the power of the distributed energy storage device in the charging direction and the discharging direction. Based on the current real-time operating status of the distributed energy storage device, the state of charge and operating temperature of the distributed energy storage device are statistically analyzed in real time to determine the safe operating power range of the distributed energy storage device. The initial power adjustment amount in the refined power adjustment command is compared with the upper power limit and the safe operating power range to determine the deviation. When the initial power adjustment exceeds the upper limit of the power or exceeds the safe operating power range, the fine-grained power adjustment command of the distributed energy storage device will be marked as an over-limit command, and the corresponding device will be marked as an over-limit device. The refined power adjustment command of the over-limit equipment is coded and corrected, and the corrected power adjustment amount is adjusted to a range that simultaneously meets the upper limit power value and the safe operating power range, thus obtaining the corrected power adjustment command of the over-limit equipment; The corrected power regulation command is encapsulated and merged with the refined power adjustment command of the device that has not been marked as exceeding the limit, forming the target coordination command for the large-scale user-side energy storage cluster.

[0014] A method for real-time coordinated control of large-scale user-side distributed energy storage clusters, based on the aforementioned system, includes the following steps: S1. Based on the real-time operating status data of distributed energy storage devices and the electrical quantity data of grid connection points, the distributed energy storage devices are dynamically grouped to obtain the energy storage subgroups of distributed energy storage devices. S2. Based on the grid dispatch requirements and the division results of the energy storage subgroup, conduct a consistency assessment of the total power demand of the energy storage subgroup, and determine the grid dispatch instructions for the energy storage subgroup based on the assessment results and the equipment parameters of the energy storage subgroup. S3. Generate a global coordinated control target for the energy storage subgroup based on the grid dispatch instructions and the local voltage-frequency over-limit events of the energy storage subgroup; S4. Based on the global collaborative control objective, distributed collaborative optimization is performed on the edge collaborative nodes of the energy storage subgroup to obtain refined power adjustment commands for the distributed energy storage devices. S5. Based on the maximum chargeable and dischargeable power of the distributed energy storage device, perform a safety verification between the fine-grained power adjustment command and the current real-time operating status of the distributed energy storage device to obtain the target coordination command for the large-scale user-side energy storage cluster.

[0015] Compared with the prior art, the present invention has the following beneficial effects: This invention dynamically clusters distributed energy storage devices based on real-time operational status data and grid-connected electrical quantity data. It constructs a voltage sensitivity matrix by correlating device state of charge with available power capacity, and forms energy storage subgroups through clustering iteration and statistical aggregation of response characteristic parameters. This accurately reveals the operational characteristics and aggregated response capabilities of each energy storage device, ensuring a high degree of alignment between the subgroup division and the actual operational status of the devices. Simultaneously, it assesses the consistency of the total power demand of the energy storage subgroups in conjunction with grid dispatch requirements. By integrating device response speed levels and response duration characteristics, it generates dispatch instructions through power correlation encoding, ensuring that the instruction content fully matches the regulation capabilities and operational characteristics of the energy storage subgroups. This significantly improves the accuracy and adaptability of grid dispatch instruction generation and enhances the responsiveness of the energy storage clusters to grid power regulation needs.

[0016] This invention combines grid dispatch instructions with local voltage-frequency over-limit events to generate a global collaborative control target. By correlating and comparing these targets, it determines control priorities and completes level matching and merging, ensuring the control target has global coordination capabilities and accurately balances grid dispatch needs with local electrical anomaly handling requirements. Based on edge collaborative nodes, it performs distributed collaborative optimization on energy storage subgroups. Through iterative interactive adjustments, it achieves collaborative convergence of regulation components in terms of power matching and response consistency. The generated refined power adjustment instructions accurately adapt to the operating status of each distributed energy storage device. Furthermore, it combines the device's maximum charge / discharge power with real-time operating status for multi-dimensional safety verification, corrects over-limit instructions, and encapsulates and merges them into target collaborative instructions. This comprehensively improves the precision of energy storage cluster control while ensuring all control instructions are executed within the safe operating range of the devices, significantly enhancing the overall accuracy and operational safety of real-time collaborative control of large-scale user-side distributed energy storage clusters. Attached Figure Description

[0017] Figure 1This is a schematic diagram of the modules of the system of the present invention; Figure 2 The power response curve of a distributed energy storage device within an energy storage subgroup after receiving a fine-tuned power adjustment command. Figure 3 The graph shows the state of charge and operating temperature changes of a distributed energy storage device within an energy storage subgroup during the power regulation process. Figure 4 This is a flowchart illustrating the method of the present invention. Detailed Implementation

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0019] Example 1: As Figure 1 As shown, the large-scale user-side distributed energy storage cluster real-time collaborative control system includes the following modules connected in sequence: The distributed energy storage dynamic clustering module is used to dynamically cluster distributed energy storage devices based on real-time operating status data and grid connection point electrical quantity data, thereby obtaining energy storage subgroups of distributed energy storage devices. The dispatch instruction generation module is used to perform a consistency assessment of the total power demand of the energy storage subgroup based on the grid dispatch requirements and the division results of the energy storage subgroup, and determine the grid dispatch instructions for the energy storage subgroup based on the assessment results and the equipment parameters of the energy storage subgroup. The global coordinated control target generation module is used to generate global coordinated control targets for the energy storage subgroup based on grid dispatch instructions and local voltage-frequency over-limit events of the energy storage subgroup. The refined power regulation module is used to perform distributed collaborative optimization on the edge collaborative nodes of the energy storage subgroup based on the global collaborative regulation target, so as to obtain the refined power adjustment command of the distributed energy storage device. The target coordination instruction output module is used to perform a safety verification between the fine-grained power adjustment instruction and the current real-time operating status of the distributed energy storage device based on the maximum chargeable and dischargeable power of the distributed energy storage device, so as to obtain the target coordination instruction of the large-scale user-side energy storage cluster.

[0020] In the distributed energy storage dynamic clustering module, the process of obtaining the energy storage subgroups of distributed energy storage devices is as follows: Based on real-time operating status data and grid connection point electrical quantity data, determine the state of charge range and voltage level of the distributed energy storage equipment; Based on the state of charge range and voltage level, a correlation analysis is performed on the current state of charge and available power capacity of distributed energy storage devices to construct a voltage sensitivity matrix for distributed energy storage devices. Clustering iteration is performed on the eigenvectors in the voltage sensitivity matrix, and distributed energy storage devices whose eigenvectors are close in distance in the feature space are grouped into the same set of devices to obtain the initial set of distributed energy storage devices. The current state of charge and available power capacity of each device in the initial device set are statistically evaluated to obtain the aggregate response characteristic parameters of the initial device set; Assign a unique virtual energy storage identifier to the initial set of devices with aggregated response characteristic parameters to obtain the energy storage subgroup of distributed energy storage devices.

[0021] Extract the state of charge (SOC) values ​​from the real-time operating status data of the distributed energy storage device. Preset SOC threshold intervals are defined as: [0, 20%], (20%, 40%], (40%, 60%], (60%, 80%], and (80%, 100%). The right endpoint of each interval is included, and interval boundaries are assigned to a lower interval. Match the SOC values ​​to the corresponding preset intervals to determine the SOC interval of the distributed energy storage device. Extract the actual grid connection voltage value from the grid connection point electrical quantity data and define... Indicates the rated voltage of the power grid, in Based on this, preset threshold values ​​for voltage grading within a given range are: Level 1 Level 2 Level 3 The boundary voltage value is classified into the superior voltage level, i.e. , Classified as Level 1 , Classified as Level 2, the actual voltage value at the grid connection point is matched to the corresponding preset level to determine the voltage level of the distributed energy storage device.

[0022] Extract the specific values ​​of the current state of charge (SOC) and available power capacity of each distributed energy storage device. Using the established SOC range and voltage level as the correlation analysis dimensions, distributed energy storage devices with the same SOC range and voltage level are grouped into the same analysis group. For each analysis group, analyze the correspondence between the device's SOC change and the grid connection point voltage change, as well as the correspondence between the device's available power capacity output and the grid connection point voltage change. Using the distributed energy storage devices as the row dimension and the voltage change-related SOC and available power capacity as the column dimension, fill the corresponding values ​​of each device into the corresponding positions in the matrix to complete the construction of the voltage sensitivity matrix of the distributed energy storage devices.

[0023] The feature vectors of distributed energy storage devices corresponding to each row of the voltage sensitivity matrix are extracted. A preset feature space distance judgment threshold is set. The position of each distributed energy storage device is marked in the feature space using the values ​​of each dimension of the feature vector as coordinates. The actual distance between the feature vectors of any two devices in the feature space is calculated. Distributed energy storage devices whose actual distance is less than or equal to the preset feature space distance judgment threshold are grouped into the same device set. If the device set ownership status changes after the first round of classification, the actual distance between the feature vectors of devices within and between sets after the change is recalculated and classified again until the ownership status of all distributed energy storage devices no longer changes. The clustering iteration of feature vectors is completed, and the final device set is the initial device set of distributed energy storage devices.

[0024] For each initial device set, the specific value of the current state of charge (SBC) of each device in the set is extracted. The sum and average value of the current SBC of all devices in the initial device set are calculated. The specific value of the available power capacity of each device in the set is extracted. The sum and maximum value of the available power capacity of all devices in the initial device set are calculated. At the same time, the number of devices with charging response capability and discharging response capability in the initial device set is counted. The sum of the SBC, the average SBC, the sum of the available power capacity, the maximum available power capacity, and the number of devices with charging and discharging response capability in the initial device set are integrated into a set of parameters. This set of parameters is the aggregate response characteristic parameter of the initial device set.

[0025] For each initial device set that has been bound to the corresponding aggregate response characteristic parameters, a unique code is assigned to each initial device set as a unique virtual energy storage identifier according to the preset character and number combination encoding rules. The virtual energy storage identifier is then bound one-to-one with the corresponding initial device set and aggregate response characteristic parameters. Each initial device set that has completed the binding operation is an independent energy storage subgroup of distributed energy storage devices.

[0026] In the dispatch instruction generation module, the process of determining the grid dispatch instructions for the energy storage subgroup is as follows: Obtain the power grid dispatching requirements issued by the superior dispatching platform. The power grid dispatching requirements include the total regional power regulation within the target time window. Based on the division results of the energy storage subgroups, the available power capacity of the distributed energy storage devices in the energy storage subgroups is statistically analyzed to obtain the adjustable power range of the energy storage subgroups. The total regional power regulation is matched with the adjustable power range of the energy storage subgroups to determine the power regulation share that each energy storage subgroup needs to undertake, and the sum of the power regulation shares of the energy storage subgroups is kept consistent with the total regional power regulation. Extract the response speed level and response duration characteristics of each energy storage subgroup from the equipment parameters of the energy storage subgroup; Power correlation encoding is performed on the characteristics of power regulation share, response speed level, and response duration to obtain the grid dispatch instructions for the energy storage subgroup. The grid dispatch instructions include the power regulation direction and power regulation amount that the energy storage subgroup needs to perform.

[0027] The power correlation coding of power regulation share, response speed level, and response duration characteristics is specifically as follows: A synergistic analysis of the power injection and power consumption demands of the energy storage subgroup within the power regulation share is conducted to determine the power regulation direction of the energy storage subgroup. Based on the amplitude characteristics of the power regulation share, the power regulation demand of the energy storage subgroup is converted accordingly to obtain the power regulation level of the power regulation share. Based on the response speed level, the power regulation response speed attribute in the energy storage subgroup is labeled and analyzed to obtain the response mode identifier of the energy storage subgroup. Based on the response duration characteristics, the power support period level of the energy storage subgroup is determined. The power support period level is used to characterize the duration for which the energy storage subgroup can maintain power output. By associating and combining the power regulation direction, power regulation level, response mode identifier, and power support time period, the grid dispatch instructions for the energy storage subgroup are obtained.

[0028] The process of obtaining the power regulation level of the power regulation share is as follows: Extract the power regulation share value of each energy storage subgroup from the power regulation share, and extract the maximum adjustable power capacity of each energy storage subgroup from the equipment parameters of the energy storage subgroup. Obtain the total number of pre-set power regulation levels, which is determined based on the regulation accuracy requirements of the energy storage cluster; Based on the power regulation share, maximum adjustable power capacity, and total number of levels, the power regulation level of the energy storage subgroup is calculated using the following formula: ; The power regulation level for the i-th energy storage subgroup, with a value range of... , Let i be the power regulation share value of the i-th energy storage subgroup. Let N be the maximum adjustable power capacity of the i-th energy storage subgroup, and N be the total number of preset power regulation levels. For floor operations, This indicates that the upper limit of the level is truncated to prevent an overflow level of N+1.

[0029] The power regulation share value is generated by the dispatch instruction generation module after executing the grid demand matching process. The upper-level dispatch platform issues the total regional power regulation amount within the target time window. The system first summarizes the adjustable power capacity of all energy storage subgroups to obtain the total adjustable upper limit of the cluster. The basic regulation share is allocated according to the proportion of the adjustable capacity of each subgroup to the total adjustable upper limit of the cluster. Then, the difference between the sum of the shares and the total regional regulation amount is adjusted and finally the power regulation share value of the corresponding energy storage subgroup is output.

[0030] The maximum adjustable power capacity is taken from the factory-calibrated hardware parameters of the distributed energy storage equipment. After the distributed energy storage dynamic clustering module completes the clustering of equipment to form energy storage subgroups, it traverses the rated charging and discharging power values ​​of all equipment in the subgroup, summarizes and statistically analyzes the absolute maximum values ​​of the maximum charging power and maximum discharging power of a single equipment, and obtains the maximum adjustable power capacity corresponding to the energy storage subgroup.

[0031] The total number of power regulation levels is generated by a fixed conversion process from the minimum regulation granularity of the power grid and the maximum adjustable power capacity of the energy storage subgroup. The system reads the minimum regulation granularity value of the power grid dispatch in the parameter configuration table, divides the maximum adjustable power capacity of the energy storage subgroup by the granularity value, and the calculation result is directly used as the total number of power regulation levels. The granularity value is explicitly specified in the regional power grid dispatch operation specification, and the capacity value is taken from the hardware statistics results of the energy storage subgroup.

[0032] The execution process of absolute value operation is to read the positive and negative signs of the power regulation share value. If the value is marked as a charging consumption condition, the negative sign is discarded and only the value is retained. If the value is marked as a discharging injection condition, the original value is directly used. The operation produces a pure amplitude value without positive or negative attributes for subsequent calculations.

[0033] The amplitude value is divided by the maximum adjustable power capacity to obtain the dimensionless capacity ratio value. This ratio value is multiplied by the total number of power adjustment levels to obtain an intermediate calculated value. The integer part of the intermediate calculated value is read down and all decimal parts are discarded. After the rounding is completed, the integer part is incremented by one to obtain the value of the level to be determined.

[0034] The upper limit truncation operation reads two values: the value of the level to be determined and the total number of power regulation levels. It compares the two values. If the value of the level to be determined is greater than the total number of power regulation levels, the total number of levels is used as the final power regulation level. If the value of the level to be determined is less than or equal to the total number of power regulation levels, the value of the level to be determined is directly used as the final power regulation level.

[0035] The power regulation level is the only output product of the entire calculation process. The value of this product strictly falls within the range of one to the total number of power regulation levels, and no invalid values ​​outside the range are generated.

[0036] The power regulation level is used to fill the hierarchical identifier field in the dispatch instruction coding process. The dispatch instruction generation module combines this level with the power regulation direction code, response mode identifier, power support time period level, original power regulation share precision value, and regulation response start and end time period in a fixed order to generate a complete power grid dispatch combination code.

[0037] After the power grid dispatch combined code is transmitted to the global coordinated control target generation module, the global coordinated control target generation module decodes and extracts the power regulation level. Based on the capacity range division rules corresponding to the level, it completes the priority ranking of the power regulation needs of each energy storage subgroup, with higher numerical levels corresponding to higher control priority sequence positions.

[0038] The regulation priority sequence is used to distinguish the order of processing grid dispatching needs and local voltage and frequency over-limit events. The power regulation needs of the energy storage subgroups that rank higher in the sequence are given priority in participating in the level matching and merging calculation of the global coordinated regulation target.

[0039] Establish a real-time communication link with the superior dispatching platform, receive power grid dispatching demand data packets issued by the platform in accordance with the standard data transmission protocol in the power dispatching field, perform structured parsing on the data packets, extract the target time window information defined in a fixed format of year-month-day-hour-minute, which clarifies the specific execution period of power regulation operations, and extract the specific value of the total regional power regulation based on active power in the data packets, so as to obtain the complete power grid dispatching demand including the total regional power regulation within the target time window.

[0040] The energy storage subgroup division results output by the distributed energy storage dynamic clustering module are retrieved. The specific values ​​of the available power capacity of each distributed energy storage device in each subgroup are extracted. The available power capacity of all devices in each energy storage subgroup is summed to obtain the total available power capacity of each energy storage subgroup. At the same time, the minimum available power capacity and the maximum available power capacity of a single device in each energy storage subgroup are determined. With the minimum available power capacity of each energy storage subgroup as the lower limit and the total available power capacity as the upper limit, the power adjustment range of each energy storage subgroup is defined. This range is the adjustable power range of the energy storage subgroup.

[0041] First, the upper limit of the adjustable power range of all energy storage subgroups is summarized to obtain the total adjustable power capacity of the large-scale user-side energy storage cluster. It is confirmed that the total adjustable power capacity is greater than or equal to the total regional power regulation. Then, the proportion of the upper limit of the adjustable power range of each energy storage subgroup to the total adjustable power capacity is used as the basis for allocation. The initial power regulation share value corresponding to each energy storage subgroup is calculated. After the initial allocation is completed, the initial power regulation share values ​​of all energy storage subgroups are summed and verified. If there is a deviation between the summation result and the total regional power regulation, the deviation value is evenly distributed to the initial power regulation share of the corresponding energy storage subgroup according to the basic allocation ratio of each energy storage subgroup. The summation and verification are carried out again until the sum of the power regulation shares of all energy storage subgroups is completely equal to the total regional power regulation. Finally, the power regulation share to be undertaken by each energy storage subgroup is determined.

[0042] The system retrieves the pre-stored equipment parameter database for each energy storage subgroup. This database records response speed-related parameters and continuous power output-related parameters for all distributed energy storage devices within the cluster. Based on a preset response speed classification standard, the response time from receiving a dispatch command to completing a power adjustment action is classified into left-closed and right-open intervals: Level 1... Level 2 Level 3 Level 4 The boundary time point is assigned to a higher response level. Response times of exactly 5s, 10s, and 20s are assigned to Level 2, Level 3, and Level 4, respectively. The average response time of all devices within each energy storage subgroup is calculated and matched to the corresponding grading standard range to determine the response speed level of each energy storage subgroup. At the same time, the maximum duration for which each energy storage subgroup can continuously output power under full-power regulation is extracted. This value is determined in conjunction with the state of charge and rated power output characteristics of the devices within the subgroup. This maximum duration value and the corresponding power output steady-state description together constitute the response duration characteristics of the energy storage subgroup.

[0043] First, determine the overall regulation attribute of the total regional power regulation. If the total regional power regulation is the power value that the grid side needs to absorb, then determine that the power regulation direction of all energy storage subgroups is charging. If the total regional power regulation is the power value that the grid side needs to release, then determine that the power regulation direction of all energy storage subgroups is discharging. Take the determined power regulation share value of each energy storage subgroup as the core power regulation amount of that subgroup. Then, according to the fixed coding format of "power regulation direction code - power regulation amount level - response mode identifier - power support period level - original power regulation share accurate value - regulation response start and end period", the six pieces of information of each energy storage subgroup are combined in an orderly manner to generate a unique code data for each energy storage subgroup. This code data is the grid dispatch instruction of the energy storage subgroup.

[0044] The power regulation share of each energy storage subgroup and the corresponding grid-side demand correlation information are retrieved. Power demand type determination rules are preset. If the power regulation share of the energy storage subgroup points to the transmission of active power to the grid, it is determined to be a power injection demand, and the corresponding power regulation direction is discharge. If the power regulation share of the energy storage subgroup points to the absorption of active power from the grid, it is determined to be a power consumption demand, and the corresponding power regulation direction is charging. The power injection demand and power consumption demand of each energy storage subgroup are checked one by one in combination with the total regional power regulation demand type of the grid dispatch. After confirming that there is no deviation, the power regulation direction of each energy storage subgroup is determined.

[0045] Extract the specific values ​​of the power regulation share of each energy storage subgroup, and simultaneously retrieve the specific values ​​of the maximum adjustable power capacity pre-stored in the equipment parameters of each energy storage subgroup. The total number of power regulation levels N is calculated by converting the minimum regulation granularity of the power grid with the maximum adjustable capacity of the subgroup. Taking N=5 as an example, the dimensionless amplitude percentage range is defined as follows: [0,20%) is Level 1, [20%,40%) is Level 2, [40%,60%) is Level 3, [60%,80%) is Level 4, and [80%,100%] is Level 5. In actual engineering, the 0~100% capacity range is evenly divided according to the calculated N value to complete the corresponding conversion of the power regulation demand of the energy storage subgroup and obtain the power regulation level of the power regulation share.

[0046] The established response speed levels of each energy storage subgroup are retrieved. These levels are divided into four levels based on response time: Level 1, Level 2, Level 3, and Level 4. A unique matching rule is preset between the response speed level and the response mode identifier. The response speed level of Level 1 corresponds to the response mode identifier V1, Level 2 corresponds to the response mode identifier V2, Level 3 corresponds to the response mode identifier V3, and Level 4 corresponds to the response mode identifier V4. The response speed levels of each energy storage subgroup are matched and marked one by one according to the matching rule, thereby completing the accurate analysis of the power regulation response speed attribute in the energy storage subgroup and obtaining the response mode identifier of the energy storage subgroup.

[0047] Extract the specific value of the longest time that the energy storage subgroup can maintain power output under full power regulation state from the response duration characteristics of each energy storage subgroup. Preset the power support period level division rules: 0 to 30 minutes is level 1, 30 to 60 minutes is level 2, 60 to 120 minutes is level 3, and more than 120 minutes is level 4. This level directly represents the attribute of how long the energy storage subgroup can maintain power output. Accurately match the longest duration value of each energy storage subgroup to the corresponding level range to determine the power support period level of the energy storage subgroup.

[0048] A unique character code is preset for the power regulation direction, with code F for discharging and code C for charging. The power regulation level, response mode identifier, and power support period are presented in standardized numerical or character form. These six pieces of information are sequentially associated and combined in a fixed order: "Power Regulation Direction Code - Power Regulation Level - Response Mode Identifier - Power Support Period - Original Power Regulation Share Precision Value - Regulation Response Start and End Time Period." This generates a unique combination code for each energy storage subgroup, which serves as the grid dispatch instruction for the subgroup. The original power regulation share precision value records the complete power regulation value allocated by the grid to the energy storage subgroup; the regulation response start and end time period records the target execution start and end times for this power regulation; and the power regulation level, response mode identifier, and power support period level are used for rapid matching and filtering of devices within the cluster.

[0049] Retrieve the relevant data on the determined power regulation share of each energy storage subgroup, and accurately extract the specific value of the power regulation share corresponding to each energy storage subgroup from the data. This value is the specific amount of power regulation that the energy storage subgroup needs to perform. At the same time, retrieve the equipment parameter database of the energy storage subgroup. This database pre-stores the equipment configuration and operating limit parameters of each energy storage subgroup. Extract the specific value of the maximum adjustable power capacity of each energy storage subgroup from the database one by one. This value is the upper limit of the adjustable power of the energy storage subgroup within the safe operating range.

[0050] The pre-stored parameter configuration table in the energy storage cluster control system is retrieved. This table stores two types of basic dimensional parameters: the minimum adjustment granularity of grid dispatch power and the rated maximum adjustable capacity of the energy storage subgroup. Based on these two types of parameters, the total number of dimensionless power adjustment levels is calculated. The conversion logic is as follows: In the formula This represents the minimum granularity of power regulation in power grid dispatching. This represents the maximum adjustable power capacity of the energy storage subgroup. It is the minimum unit of power regulation required by grid dispatch. MW, maximum adjustable power of subgroup MW, substituting into the calculation, we get the total number of power regulation levels N=10; if the minimum power regulation unit required by the power grid dispatching... MW, maximum adjustable power of subgroup Substituting MW into the calculation yields a total of N=5 power regulation levels; the basic parameters are directly read from the parameter configuration table to complete the conversion, and the specific value of the total number of power regulation levels adapted to the current energy storage subgroup and scheduling accuracy requirements is extracted.

[0051] For each energy storage subgroup, its corresponding power regulation share value is divided by the maximum adjustable power capacity to obtain the ratio. This ratio is then multiplied by the pre-set total number of power regulation levels to obtain the multiplied value. This multiplied value is then rounded down to the nearest integer. If the multiplied value is an integer, it is taken directly; if it is a decimal, the decimal part is discarded and only the integer part is retained. Finally, 1 is added to the rounded value to obtain the final value, which is the power regulation level of the energy storage subgroup.

[0052] From the established power regulation share data of the energy storage subgroups, the specific value of the power regulation share corresponding to each energy storage subgroup is accurately extracted. This value represents the specific amount of power regulation that the energy storage subgroup needs to perform.

[0053] The maximum adjustable power capacity of each energy storage subgroup is extracted one by one from the pre-stored equipment parameter database. This value is the upper limit of the adjustable power of the energy storage subgroup within the safe operating range.

[0054] The total number of power regulation levels is extracted from the parameter configuration table pre-stored in the energy storage cluster control system. This value is a fixed value pre-set according to the grid dispatch control accuracy requirements of the energy storage cluster.

[0055] Rounding down is a fixed operation rule applied to the results generated during numerical calculations. This rule takes the largest integer not greater than the calculation result and is a predetermined operation method used in calculating the power regulation level.

[0056] The power regulation level is the numerical result of the classification of the energy storage subgroup after calculating the power regulation share, maximum adjustable power capacity, and total number of power regulation levels of the energy storage subgroup in a fixed manner.

[0057] This calculation method can correlate the actual power regulation requirements of the energy storage subgroup with its maximum adjustable power capacity. Combined with the total number of power regulation levels set in advance according to the control accuracy, it can achieve standardized hierarchical quantification of the power regulation requirements of the energy storage subgroup through fixed calculation steps.

[0058] This calculation method ensures that the generated power regulation levels accurately reflect the proportion of power regulation share of each energy storage subgroup to its maximum adjustable power capacity, thus ensuring that the classification results match the actual regulation capabilities of the energy storage subgroups.

[0059] This calculation method assigns a unified standard level label to the power regulation requirements of different energy storage subgroups through a standardized calculation process, avoiding the problem of inconsistent quantitative standards for regulation requirements caused by differences in the regulation capabilities of the energy storage subgroups themselves.

[0060] The power regulation level generated by this calculation method serves as the core quantitative indicator for power correlation coding, providing a precise and unified hierarchical basis for the standardized generation of subsequent power grid dispatch instructions, and ensuring that dispatch instructions can accurately reflect the power regulation demand levels of each energy storage subgroup.

[0061] This calculation method uses fixed operation rules to ensure that each energy storage subgroup can obtain a unique corresponding power regulation level, ensuring that the generation of power regulation levels is reproducible and that the power regulation and tiered operation of energy storage clusters has a unified execution standard.

[0062] In the global coordinated regulation target generation module, the process of generating the global coordinated regulation target for the energy storage subgroup is as follows: The over-limit type, over-limit magnitude, and over-limit duration of each energy storage subgroup are aggregated into local voltage-frequency over-limit event information for the energy storage subgroup; The power grid dispatching command is decoded and separated to obtain the power regulation requirements of the power grid dispatching command. The power regulation requirements include the power regulation direction, power regulation amount and regulation response time period. The power regulation requirements are correlated and compared with local voltage-frequency over-limit event information to obtain the regulation-event correlation degree identifier of the energy storage subgroup; Based on the regulation-event correlation identifier, the power regulation requirements of each subgroup in the energy storage subgroup are prioritized to obtain the regulation priority sequence of power regulation requirements; The power regulation requirements of different energy storage subgroups in the regulation priority sequence are matched and merged to obtain the global coordinated regulation target of the energy storage subgroups. The global coordinated regulation target includes the total power regulation amount and the overall response time window.

[0063] Electrical anomaly data during the operation of each energy storage subgroup is collected. The over-limit type is identified as either voltage over-limit or frequency over-limit. The difference between the actual voltage value and the grid rated voltage value for voltage over-limit, and the difference between the actual frequency value and the grid rated frequency value for frequency over-limit are extracted as the over-limit amplitude. The time length from the time the over-limit state occurs to the current data collection time is counted by the timing module as the over-limit duration. The above three types of information from each energy storage subgroup are uniformly collected and organized according to a fixed structured format of "energy storage subgroup identifier-over-limit type-over-limit amplitude-over-limit duration" to form the local voltage-frequency over-limit event information of the energy storage subgroup.

[0064] The grid dispatch instruction for the energy storage subgroup output by the dispatch instruction generation module is retrieved. This instruction is in a standardized combined coding form. According to the preset grid dispatch instruction decoding rules and the preset segmentation rules, the grid dispatch combined coding is decoded and separated, and six items are parsed in sequence: power regulation direction code, power regulation amount level, response mode identifier, power support time period level, original power regulation share precision value, and regulation response start and end time period. These are integrated to form a complete power regulation requirement. The power regulation requirement includes three core contents: power regulation direction, precise power regulation amount, and regulation response time period. The power regulation direction is obtained by parsing the power regulation direction code. The precise power regulation amount is directly read from the original power regulation share precision value stored in the code. The regulation response time period is directly read from the regulation response start and end time period stored in the code. At the same time, the power regulation amount level and power support time period level are extracted as auxiliary parameters to participate in the subsequent control priority determination. The information obtained after splitting and parsing is integrated to obtain the power regulation requirement of the grid dispatch instruction containing the power regulation direction, power regulation amount, and regulation response time period.

[0065] The pre-defined regulation-event correlation determination rule is based on three dimensions: the time overlap between the regulation response period of the power regulation requirement and the over-limit duration of the local voltage-frequency over-limit event information; the adaptability between the power regulation amount of the power regulation requirement and the over-limit amplitude; and the matching degree between the power regulation direction and the governance requirements of the over-limit type. If all three dimensions match, it is determined as high correlation and assigned the label G1; if two dimensions match, it is determined as medium correlation and assigned the label G2; and if one or less dimensions match, it is determined as low correlation and assigned the label G3. For each subgroup, the power regulation requirement and the local voltage-frequency over-limit event information are compared and correlated item by item according to this rule. The corresponding label is determined based on the comparison result. This label is the regulation-event correlation label of the energy storage subgroup.

[0066] The preset regulation-event correlation identifier corresponds to the regulation priority rule. Identifier G1 corresponds to the first-level regulation priority, identifier G2 corresponds to the second-level regulation priority, and identifier G3 corresponds to the third-level regulation priority. The first-level regulation priority is higher than the second-level regulation priority, and the second-level regulation priority is higher than the third-level regulation priority. Energy storage subgroups with the same level of regulation priority are arranged in descending order of the over-limit amplitude. According to this rule, the power regulation requirements of each subgroup in the energy storage subgroup are arranged in an orderly manner to form an orderly power regulation requirement regulation priority sequence.

[0067] Extract the specific values ​​of power regulation within the power regulation requirements of all energy storage subgroups in the regulation priority sequence, sum all the values, and the result is the total power regulation in the global coordinated regulation target. Extract the earliest start time and the latest end time of the regulation response period of each energy storage subgroup in the regulation priority sequence, define this time interval as the overall response time window in the global coordinated regulation target, and integrate and encapsulate the total power regulation and the overall response time window to form the global coordinated regulation target of the energy storage subgroup that includes these two core contents.

[0068] In the refined power control module, the process of obtaining refined power adjustment commands from distributed energy storage devices is as follows: It receives global coordinated control targets and obtains the current real-time operating status of each distributed energy storage device in the energy storage subgroup; Based on the edge collaborative nodes of the energy storage subgroup, combined with the global collaborative control target and the current real-time operating status, the initial power regulation requirement of the energy storage subgroup is obtained, and the initial power regulation requirement is decomposed into preliminary regulation components corresponding to each distributed energy storage device. The initial adjustment components are sent to the corresponding distributed energy storage devices, and the acceptance feedback information returned by the distributed energy storage devices through their own real-time constraints is received. Based on the feedback information, the initial adjustment components are iteratively adjusted until the adjustment components of the distributed energy storage devices reach a preset collaborative convergence state in terms of power matching degree and response consistency. The process is as follows: Read the aggregated response characteristic parameters of the distributed energy storage device and the cluster collaborative control requirements of the edge collaborative nodes respectively; The edge collaboration node interacts and adjusts with each distributed energy storage device in the energy storage subgroup to obtain the current state of charge, adjustability and local electrical safety boundary information of the distributed energy storage device; Based on the requirements for cluster coordinated control and combined with the aggregation response characteristics of the energy storage subgroup, the power sharing intentions of the equipment in the energy storage subgroup are initially proposed, and the initial power sharing intention set of the energy storage subgroup is obtained. The initial power sharing intention set is exchanged with the edge collaborative nodes of the adjacent energy storage subgroups, and the power sharing intention information of the adjacent energy storage subgroups is received. Based on the power sharing intention information, load judgment is made on the local network overload and equipment charge status of large-scale user-side energy storage clusters, and the power sharing intention is adaptively adjusted according to the judgment results. Each adjustment component after reaching the collaborative convergence state is determined as a fine-grained power adjustment command for each distributed energy storage device, and the fine-grained power adjustment command is sent to the corresponding distributed energy storage device.

[0069] Through a dedicated power data communication link established with the global coordinated control target generation module, the complete data packet of the global coordinated control target is received in accordance with the standardized data encapsulation format of the power system. The data packet is parsed to extract two core contents: the total power regulation amount and the overall response time window. At the same time, through the real-time monitoring sensor units on each distributed energy storage device, data such as the current state of charge, real-time charging and discharging power, device operating temperature, and available power capacity of the device are acquired at a sampling frequency of once per second. All collected data are classified and summarized according to the unique device identifier to ensure that the timestamp of all data is consistent with the current collection time, forming the current real-time operating status of each distributed energy storage device in the energy storage subgroup.

[0070] Each energy storage subgroup's edge coordination node acts as the local distributed control core node, pre-storing the basic operating constraint parameters of all distributed energy storage devices within the subgroup. This node retrieves the received global coordinated control targets and the summarized real-time operating status of the devices. First, based on the overall response time window in the global coordinated control targets, it determines the unified execution period for power regulation of all devices within the subgroup. Then, combining the ratio of the total overall power regulation to the total available power capacity of all devices within the subgroup, it determines the specific power regulation value that the energy storage subgroup needs to undertake. This value is the initial power regulation requirement of the energy storage subgroup. Subsequently, according to the ratio of the available power capacity of each distributed energy storage device to the total available power capacity of the subgroup, the initial power regulation requirement of the energy storage subgroup is allocated one by one to each distributed energy storage device within the subgroup. After allocation, the allocated values ​​of all devices are summed and verified to ensure that the summation result is completely consistent with the initial power regulation requirement, forming a preliminary regulation component corresponding to each distributed energy storage device.

[0071] The edge collaboration nodes of the energy storage sub-group accurately send preliminary adjustment components carrying unique device identifiers to the corresponding distributed energy storage devices via an industrial Ethernet communication bus. The sent preliminary adjustment components include three core pieces of information: power adjustment direction, adjustment amount, and adjustment execution period. The local control unit of each distributed energy storage device retrieves its own preset real-time constraint parameters, which include the device's maximum charge and discharge power, upper and lower limits of state of charge, and safe operating temperature range. The received preliminary adjustment components are compared item by item with its own real-time constraint parameters. If the preliminary adjustment components are completely within the range of real-time constraint parameters, the local control unit generates an acceptance feedback message marked "accept" along with the device's current real-time operating parameters. If the preliminary adjustment components exceed the range of real-time constraint parameters, the local control unit generates an acceptance feedback message marked "reject" along with the specific over-limit parameters and the adjustment amount range that the device can accept. All devices transmit their acceptance feedback information back to the edge collaboration nodes via the original communication link. The nodes then summarize all acceptance feedback information according to the unique device identifier.

[0072] A pre-defined criterion for determining the convergence state of distributed collaborative optimization is established. This criterion includes three conditions that must be met simultaneously: first, all distributed energy storage devices within the energy storage subgroup must return acceptance feedback information marked as "accepted"; second, the sum of the adjustment components of all devices within the subgroup must deviate from the initial power adjustment demand of the energy storage subgroup by 0; and third, the power adjustment execution period of each device must perfectly match the overall response time window of the global collaborative control target. The edge collaborative nodes first filter out devices marked as "rejected" from the aggregated acceptance feedback information, extract the acceptable adjustment range of their feedback, and adjust the initial adjustment component of the device to its maximum acceptable adjustment value. Then, the power adjustment gap caused by this adjustment is redistributed according to the proportion of the remaining available power capacity of the devices marked as "accepted". After completing one adjustment, a new adjustment component is generated and sent back to the corresponding device. The new acceptance feedback information returned by the device is received. If the new feedback information still does not meet the three criteria for determining the collaborative convergence state, the above operation of adjusting, sending, and receiving adjustment components is repeated until all acceptance feedback information from all devices is "accepted" and all three criteria are met.

[0073] After the edge collaborative nodes of the energy storage subgroup reach the preset collaborative convergence state, they will structurally encapsulate the final adjustment components corresponding to each distributed energy storage device. The encapsulation content includes the device's unique identifier, power adjustment direction, precise adjustment amount, adjustment start time, and adjustment end time. This structurally encapsulated final adjustment component is the fine-grained power adjustment instruction for each distributed energy storage device. The edge collaborative nodes will then use an encrypted industrial communication link to precisely resend the fine-grained power adjustment instruction for each device to the corresponding distributed energy storage device according to the device's unique identifier. After the resending is completed, an instruction resending record will be automatically generated. The record content includes the instruction resending time, the device's unique identifier, and the complete content of the fine-grained power adjustment instruction. At the same time, each distributed energy storage device is required to immediately return a reception confirmation message containing its own identifier after receiving the instruction. The edge collaborative nodes will verify the returned reception confirmation messages one by one to ensure that all distributed energy storage devices in the subgroup have successfully received the corresponding fine-grained power adjustment instruction.

[0074] The parameter repository of the distributed energy storage dynamic clustering module is precisely retrieved according to the virtual energy storage identifier of each energy storage sub-group. The aggregate response characteristic parameters of the corresponding distributed energy storage devices are read. These parameters include core information such as the total state of charge, the total available power capacity, and the number of devices with charge and discharge response capabilities. At the same time, the cluster collaborative control requirements are read from the local control requirement storage area of ​​the edge collaborative nodes. These requirements are converted from the global collaborative control objectives and include fixed information such as the total power adjustment of the large-scale user-side energy storage cluster, the unified response time window, and the network transmission power constraint value. During the reading process, unique device and sub-group identifiers are added to both types of data to ensure that the data is accurately matched with the corresponding energy storage sub-group and distributed energy storage device.

[0075] The edge collaboration nodes of the energy storage sub-group send status interaction acquisition commands to each distributed energy storage device within the sub-group via a low-latency industrial communication link. The commands carry acquisition timestamps and data feedback format requirements. Upon receiving the commands, the local control units of each distributed energy storage device immediately extract the specific value of its current state of charge, the adjustable capacity value calculated by combining the maximum charging and discharging power and the current operating power, and retrieve local electrical safety boundary information such as voltage safety threshold, frequency safety threshold, and operating temperature safety threshold stored in the device. The device encapsulates the above three types of information in a specified format and transmits them back to the edge collaboration nodes through the original communication link. The nodes perform integrity verification on the transmitted data. After the verification is passed, the data is classified and organized according to the unique device identifier to obtain the current state of charge, adjustable capacity, and local electrical safety boundary information of each distributed energy storage device.

[0076] The edge collaboration node divides the total power regulation in the cluster collaborative control demand into the corresponding energy storage subgroups according to the proportion of available power capacity of each energy storage subgroup. Then, combined with the aggregation response characteristics of the energy storage subgroup, it proposes a specific power sharing value for each device according to the proportion of the adjustable capacity of each distributed energy storage device to the total available power capacity of the subgroup. At the same time, combined with the unified response time window of the cluster collaborative control demand, it proposes the power sharing execution period of each device. The device's unique identifier, the proposed power sharing value, and the power sharing execution period are associated and bound together, and organized in an orderly manner according to the device number within the subgroup, forming an initial power sharing intention set of the energy storage subgroup containing the power sharing information of all distributed energy storage devices within the subgroup.

[0077] Based on the pre-stored network topology of the large-scale user-side energy storage cluster, the edge collaborative nodes determine the adjacent energy storage subgroups of their corresponding energy storage subgroups in the network. Through the cluster's dedicated collaborative communication link, they establish data transmission connections with the edge collaborative nodes of the adjacent energy storage subgroups. They encapsulate the initial power sharing intention set of their own energy storage subgroup according to a preset standardized data format, add subgroup identifiers, and send it to each adjacent edge collaborative node. At the same time, they receive the initial power sharing intention sets of their energy storage subgroups returned by each adjacent edge collaborative node. They parse and verify the integrity of the received data packets, remove invalid data, and classify and organize them according to the adjacent subgroup identifiers to obtain the power sharing intention information of the adjacent energy storage subgroups.

[0078] Edge collaborative nodes retrieve the grid topology and rated transmission capacity parameters of the large-scale user-side energy storage cluster. Combined with their own and adjacent energy storage subgroups' power sharing intention information, they calculate the estimated transmission power values ​​of each distribution line. If the estimated transmission power value exceeds the rated transmission capacity parameter of the line, the corresponding line area is determined to be locally overloaded. At the same time, they extract the estimated state of charge (SOC) values ​​of each distributed energy storage device after the proposed power sharing is executed from the power sharing intention information. A safe SOC range of 20% to 80% is preset. If the estimated value exceeds this range, the corresponding device is determined to be overloaded. For locally overloaded areas, part of the power sharing on the overloaded line is transferred to devices in adjacent subgroups with sufficient adjustability and no risk of overload on the line. For overloaded devices, their proposed power sharing value is reduced until the estimated SOC falls within the safe range. The reduced power sharing is then redistributed to devices in the same subgroup with a safe SOC and remaining adjustability. After the above adjustments are completed, a new power sharing intention is formed.

[0079] In the target coordination instruction output module, the process of obtaining the target coordination instructions for a large-scale user-side energy storage cluster is as follows: The maximum chargeable and dischargeable power is extracted from the equipment parameters of the distributed energy storage device to obtain the power limit set of the distributed energy storage device. The power limit set includes the upper limit value of the power of the distributed energy storage device in the charging direction and the discharging direction. Based on the current real-time operating status of the distributed energy storage device, the state of charge and operating temperature of the distributed energy storage device are statistically analyzed in real time to determine the safe operating power range of the distributed energy storage device. The initial power adjustment amount in the refined power adjustment command is compared with the upper power limit and the safe operating power range to determine the deviation. When the initial power adjustment exceeds the power limit or exceeds the safe operating power range, the fine-tuning power adjustment command of the distributed energy storage device will be marked as an over-limit command, and the corresponding device will be marked as an over-limit device. The refined power adjustment command of the over-limit equipment is coded and corrected, and the corrected power adjustment amount is adjusted to a range that simultaneously meets the upper limit power value and the safe operating power range, thus obtaining the corrected power adjustment command of the over-limit equipment; The corrected power regulation command is encapsulated and merged with the refined power adjustment command of the device that has not been marked as exceeding the limit, forming the target coordination command for the large-scale user-side energy storage cluster.

[0080] From the dedicated repository of device parameters for distributed energy storage devices, the rated maximum charging power and rated maximum discharging power parameters calibrated by the manufacturer are retrieved one by one according to the unique identifier of each device. These parameters are fixed operating limit values ​​for the devices without any adjustment deviation. The upper limit values ​​of the charging direction power and the upper limit values ​​of the discharging direction power for each device are associated and bound, generating a unique set of power limit related data for each device. This set of data is the power limit set for distributed energy storage devices, which only contains the upper limit values ​​of the power corresponding to the charging and discharging directions for each device.

[0081] The real-time status acquisition unit of the distributed energy storage device obtains the specific values ​​of the current state of charge and the real-time operating temperature. According to the power energy storage device operation standards, the preset state of charge ranges are 0~20%, 20%~80%, and 80%~100%, and the operating temperature ranges are 0~45℃, 45~55℃, and 55℃ and above. When the state of charge is 20%~80% and the operating temperature is 0~45℃, the safe charging and discharging power range of the device is from 0 to the upper limit of the corresponding power value. Discharging is prohibited when the state of charge is 0~20%. The safe operating power range for charging is 0% to 50% of the maximum charging power. Charging is prohibited when the state of charge is 80% to 100%, and the safe operating power range for discharging is 0% to 50% of the maximum discharging power. When the operating temperature is 45 to 55°C, the safe operating power range for both charging and discharging is 0% to 70% of the corresponding maximum power. When the operating temperature is 55°C or above, the safe operating power range for both charging and discharging is 0%. According to this corresponding rule, a unique charging and discharging power range is determined for each device, and this range is the safe operating power range for distributed energy storage devices.

[0082] The preliminary power adjustment amount contained in the refined power adjustment instruction is retrieved according to the unique identifier of each device. This adjustment amount clearly indicates the power adjustment direction and specific value. The preliminary power adjustment amount in the charging direction is compared one by one with the upper limit of charging power in the power limit set and the charging power range of the safe operating power range. The preliminary power adjustment amount in the discharging direction is compared one by one with the upper limit of discharging power in the power limit set and the discharging power range of the safe operating power range. The numerical matching results of the preliminary power adjustment amount of each device in each comparison dimension are recorded, and it is clearly marked whether the value is within the range of the corresponding standard.

[0083] The preset rule for determining if an instruction exceeds the limit is that the initial power adjustment value exceeds the upper limit of the corresponding direction of the power limit set, or the initial power adjustment value exceeds the power range of the corresponding direction of the safe operating power range. If either condition is met, the instruction is determined to be an over-limit instruction. A fixed "over-limit instruction" text identifier is added to the refined power adjustment instruction of the distributed energy storage device that is determined to be an over-limit instruction. At the same time, a fixed "over-limit device" text identifier is added to the distributed energy storage device. The over-limit instruction identifier and the over-limit device identifier are accurately associated with the device's unique identifier, and an over-limit list containing the device identifier and the corresponding instruction is compiled. Devices and instructions that do not meet the over-limit determination rule are not marked.

[0084] For each device on the over-limit list, extract the initial power adjustment direction and specific reason for exceeding the limit from its refined power adjustment instruction. If it only exceeds the power upper limit, the adjustment value is corrected to the power upper limit value in the corresponding direction of the power limit set. If it only exceeds the safe operating power range, the adjustment value is corrected to the maximum value in the corresponding direction of the safe operating power range. If it exceeds both, the adjustment value is corrected to the maximum value in the corresponding direction that simultaneously satisfies the power upper limit and the safe operating power range. During the correction process, the instruction's adjustment execution period, device unique identifier, and other content remain unchanged. The corrected adjustment amount is re-standardized and encoded according to the original refined power adjustment instruction's encoding format, and a "correction" exclusive identifier is added. This re-encoded instruction is the corrected power adjustment instruction for the over-limit device.

[0085] All relevant instructions for large-scale user-side energy storage clusters are categorized according to the energy storage subgroup identifiers. Corrective power adjustment instructions for over-limit devices and refined power adjustment instructions for devices not marked as over-limit are comprehensively summarized to ensure no device instructions are omitted or duplicated. All instructions are arranged in an orderly manner according to both subgroup identifiers and device unique identifiers. All arranged instructions are then uniformly structured and encapsulated, with the encapsulation content including the overall cluster identifier, subgroup identifier, device unique identifier, power adjustment direction, power adjustment amount, adjustment execution period, and instruction validity check code. The encapsulated overall instruction data set is packaged according to the power system dispatch standard data format. This packaged complete data set is the target coordination instruction for the large-scale user-side energy storage cluster.

[0086] The algorithm involved in this embodiment can be executed by an electronic device, which includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The algorithm calculation is implemented by the processor executing the software.

[0087] Figure 2The graph shows the power response curve of a distributed energy storage device within the energy storage subgroup after receiving a fine-grained power adjustment command. The solid blue line represents the actual output power of the device, and the dashed red line represents the response speed threshold. The device linearly ramps up from receiving the command at 0 seconds, reaching the target power of 90kW in approximately 15 seconds. Thereafter, it stabilizes around 90kW with only minor fluctuations, indicating that the device's adjustment component has achieved the preset cooperative convergence state in terms of power matching and response consistency. Its approximately 15-second ramp-up time falls within the three-level response speed range [10s, 20s] defined in this invention, verifying the precise matching between the fine-grained power adjustment command and the device's actual adjustment capability.

[0088] Figure 3 The graph shows the state of charge (SOC) and operating temperature changes during power regulation of the same device. The solid green line represents the SOC curve, and the dashed purple line represents the operating temperature curve. During the 60-second continuous discharge, the SOC slowly decreased from approximately 0.301 to approximately 0.290, reflecting the gradual consumption of stored energy with power output. The operating temperature slowly increased from approximately 32.8℃ to approximately 36℃, reflecting the heat generation process during device operation. Throughout the regulation process, the SOC remained within the safe range (20%~80%) set by the safety verification, and the operating temperature remained within the normal operating temperature range (0~45℃). This indicates that the device's refined power adjustment command passed the multi-dimensional safety verification based on the SOC and operating temperature, and the device performed power regulation within the safe operating range.

[0089] Example 2: As Figure 4 As shown, the method for real-time coordinated control of large-scale user-side distributed energy storage clusters, based on the system in Example 1, includes the following steps: S1. Based on the real-time operating status data of distributed energy storage devices and the electrical quantity data of grid connection points, the distributed energy storage devices are dynamically grouped to obtain the energy storage subgroups of distributed energy storage devices. S2. Based on the grid dispatch requirements and the division results of the energy storage subgroup, conduct a consistency assessment of the total power demand of the energy storage subgroup, and determine the grid dispatch instructions for the energy storage subgroup based on the assessment results and the equipment parameters of the energy storage subgroup. S3. Generate a global coordinated control target for the energy storage subgroup based on the grid dispatch instructions and the local voltage-frequency over-limit events of the energy storage subgroup; S4. Based on the global collaborative control objective, distributed collaborative optimization is performed on the edge collaborative nodes of the energy storage subgroup to obtain refined power adjustment commands for the distributed energy storage devices. S5. Based on the maximum chargeable and dischargeable power of the distributed energy storage device, perform a safety verification between the fine-grained power adjustment command and the current real-time operating status of the distributed energy storage device to obtain the target coordination command for the large-scale user-side energy storage cluster.

[0090] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can make equivalent substitutions or modifications based on the technical solution and concept of the present invention within the scope of the technology disclosed in the present invention, and such modifications should also be considered to fall within the scope of protection of the present invention.

Claims

1. A large-scale user-side distributed energy storage cluster real-time collaborative control system, characterized in that, Includes the following modules connected in sequence: The distributed energy storage dynamic clustering module is used to dynamically cluster distributed energy storage devices based on real-time operating status data and grid connection point electrical quantity data, thereby obtaining energy storage subgroups of distributed energy storage devices. The dispatch instruction generation module is used to perform a consistency assessment of the total power demand of the energy storage subgroup based on the grid dispatch requirements and the division results of the energy storage subgroup, and determine the grid dispatch instructions for the energy storage subgroup based on the assessment results and the equipment parameters of the energy storage subgroup. The global coordinated control target generation module is used to generate global coordinated control targets for the energy storage subgroup based on grid dispatch instructions and local voltage-frequency over-limit events of the energy storage subgroup. The refined power regulation module is used to perform distributed collaborative optimization on the edge collaborative nodes of the energy storage subgroup based on the global collaborative regulation target, so as to obtain the refined power adjustment command of the distributed energy storage device. The target coordination instruction output module is used to perform a safety verification between the fine-grained power adjustment instruction and the current real-time operating status of the distributed energy storage device based on the maximum chargeable and dischargeable power of the distributed energy storage device, so as to obtain the target coordination instruction of the large-scale user-side energy storage cluster.

2. The large-scale user-side distributed energy storage cluster real-time collaborative control system as described in claim 1, characterized in that, In the distributed energy storage dynamic clustering module, the process of obtaining the energy storage subgroups of distributed energy storage devices is as follows: Based on real-time operating status data and grid connection point electrical quantity data, determine the state of charge range and voltage level of the distributed energy storage equipment; Based on the state of charge range and voltage level, a correlation analysis is performed on the current state of charge and available power capacity of distributed energy storage devices to construct a voltage sensitivity matrix for distributed energy storage devices. Clustering iteration is performed on the eigenvectors in the voltage sensitivity matrix, and distributed energy storage devices whose eigenvectors are close in distance in the feature space are grouped into the same set of devices to obtain the initial set of distributed energy storage devices. The current state of charge and available power capacity of each device in the initial device set are statistically evaluated to obtain the aggregate response characteristic parameters of the initial device set; Assign a unique virtual energy storage identifier to the initial set of devices with aggregated response characteristic parameters to obtain the energy storage subgroup of distributed energy storage devices.

3. The large-scale user-side distributed energy storage cluster real-time collaborative control system as described in claim 1, characterized in that, In the dispatch instruction generation module, the process of determining the grid dispatch instructions for the energy storage subgroup is as follows: Obtain the power grid dispatching requirements issued by the superior dispatching platform. The power grid dispatching requirements include the total regional power regulation within the target time window. Based on the division results of the energy storage subgroups, the available power capacity of the distributed energy storage devices in the energy storage subgroups is statistically analyzed to obtain the adjustable power range of the energy storage subgroups. The total regional power regulation is matched with the adjustable power range of the energy storage subgroups to determine the power regulation share that each energy storage subgroup needs to undertake, and the sum of the power regulation shares of the energy storage subgroups is kept consistent with the total regional power regulation. Extract the response speed level and response duration characteristics of each energy storage subgroup from the equipment parameters of the energy storage subgroup; Power correlation encoding is performed on the characteristics of power regulation share, response speed level, and response duration to obtain the grid dispatch instructions for the energy storage subgroup. The grid dispatch instructions include the power regulation direction and power regulation amount that the energy storage subgroup needs to perform.

4. The large-scale user-side distributed energy storage cluster real-time collaborative control system as described in claim 3, characterized in that, In the scheduling instruction generation module, the power regulation share, response speed level, and response duration characteristics are specifically encoded using power correlation as follows: A synergistic analysis of the power injection and power consumption demands of the energy storage subgroup within the power regulation share is conducted to determine the power regulation direction of the energy storage subgroup. Based on the amplitude characteristics of the power regulation share, the power regulation demand of the energy storage subgroup is converted accordingly to obtain the power regulation level of the power regulation share. Based on the response speed level, the power regulation response speed attribute in the energy storage subgroup is labeled and analyzed to obtain the response mode identifier of the energy storage subgroup. Based on the response duration characteristics, the power support period level of the energy storage subgroup is determined. The power support period level is used to characterize the duration for which the energy storage subgroup can maintain power output. By associating and combining the power regulation direction, power regulation level, response mode identifier, and power support time period, the grid dispatch instructions for the energy storage subgroup are obtained.

5. The large-scale user-side distributed energy storage cluster real-time collaborative control system as described in claim 4, characterized in that, In the dispatch instruction generation module, the process of obtaining the power regulation level of the power regulation share is as follows: Extract the power regulation share value of each energy storage subgroup from the power regulation share, and extract the maximum adjustable power capacity of each energy storage subgroup from the equipment parameters of the energy storage subgroup. Obtain the total number of pre-set power regulation levels, which is determined based on the regulation accuracy requirements of the energy storage cluster; Based on the power regulation share, maximum adjustable power capacity, and total number of levels, the power regulation level of the energy storage subgroup is calculated using the following formula: ； The power regulation level for the i-th energy storage subgroup, with a value range of... , Let i be the power regulation share value of the i-th energy storage subgroup. Let N be the maximum adjustable power capacity of the i-th energy storage subgroup, and N be the total number of preset power regulation levels. This is for floor function.

6. The large-scale user-side distributed energy storage cluster real-time collaborative control system as described in claim 1, characterized in that, In the global coordinated regulation target generation module, the process of generating the global coordinated regulation target for the energy storage subgroup is as follows: The over-limit type, over-limit magnitude, and over-limit duration of each energy storage subgroup are aggregated into local voltage-frequency over-limit event information for the energy storage subgroup; The power grid dispatching command is decoded and separated to obtain the power regulation requirements of the power grid dispatching command. The power regulation requirements include the power regulation direction, power regulation amount and regulation response time period. The power regulation requirements are correlated and compared with local voltage-frequency over-limit event information to obtain the regulation-event correlation degree identifier of the energy storage subgroup; Based on the regulation-event correlation identifier, the power regulation requirements of each subgroup in the energy storage subgroup are prioritized to obtain the regulation priority sequence of power regulation requirements; The power regulation requirements of different energy storage subgroups in the regulation priority sequence are matched and merged to obtain the global coordinated regulation target of the energy storage subgroups. The global coordinated regulation target includes the total power regulation amount and the overall response time window.

7. The large-scale user-side distributed energy storage cluster real-time collaborative control system as described in claim 1, characterized in that, In the refined power control module, the process of obtaining refined power adjustment commands from distributed energy storage devices is as follows: It receives global coordinated control targets and obtains the current real-time operating status of each distributed energy storage device in the energy storage subgroup; Based on the edge collaborative nodes of the energy storage subgroup, combined with the global collaborative control target and the current real-time operating status, the initial power regulation requirement of the energy storage subgroup is obtained, and the initial power regulation requirement is decomposed into preliminary regulation components corresponding to each distributed energy storage device. The initial adjustment components are sent to the corresponding distributed energy storage devices, and the acceptance feedback information returned by the distributed energy storage devices through their own real-time constraints is received. Based on the feedback information, the initial adjustment components are iteratively and interactively adjusted until the adjustment components of the distributed energy storage device reach the preset collaborative convergence state in terms of power matching degree and response consistency. Each adjustment component after reaching the collaborative convergence state is determined as a fine-grained power adjustment command for each distributed energy storage device, and the fine-grained power adjustment command is sent to the corresponding distributed energy storage device.

8. The large-scale user-side distributed energy storage cluster real-time collaborative control system as described in claim 7, characterized in that, In the refined power regulation module, the initial regulation component is iteratively and interactively adjusted until the regulation component of the distributed energy storage device reaches a preset collaborative convergence state in terms of power matching degree and response consistency. Specifically: Read the aggregated response characteristic parameters of the distributed energy storage device and the cluster collaborative control requirements of the edge collaborative nodes respectively; The edge collaboration node interacts and adjusts with each distributed energy storage device in the energy storage subgroup to obtain the current state of charge, adjustability and local electrical safety boundary information of the distributed energy storage device; Based on the requirements for cluster coordinated control and combined with the aggregation response characteristics of the energy storage subgroup, the power sharing intentions of the equipment in the energy storage subgroup are initially proposed, and the initial power sharing intention set of the energy storage subgroup is obtained. The initial power sharing intention set is exchanged with the edge collaborative nodes of the adjacent energy storage subgroups, and the power sharing intention information of the adjacent energy storage subgroups is received. Based on the power sharing intention information, load judgment is made on the local network overload and equipment charge status of large-scale user-side energy storage clusters, and the power sharing intention is adaptively adjusted according to the judgment results.

9. The large-scale user-side distributed energy storage cluster real-time collaborative control system as described in claim 1, characterized in that, In the target coordination command output module, the process of obtaining the target coordination command for a large-scale user-side energy storage cluster is as follows: The maximum chargeable and dischargeable power is extracted from the equipment parameters of the distributed energy storage device to obtain the power limit set of the distributed energy storage device. The power limit set includes the upper limit value of the power of the distributed energy storage device in the charging direction and the discharging direction. Based on the current real-time operating status of the distributed energy storage device, the state of charge and operating temperature of the distributed energy storage device are statistically analyzed in real time to determine the safe operating power range of the distributed energy storage device. The initial power adjustment amount in the refined power adjustment command is compared with the upper power limit and the safe operating power range to determine the deviation. When the initial power adjustment exceeds the upper limit of the power or exceeds the safe operating power range, the fine-grained power adjustment command of the distributed energy storage device will be marked as an over-limit command, and the corresponding device will be marked as an over-limit device. The refined power adjustment command of the over-limit equipment is coded and corrected, and the corrected power adjustment amount is adjusted to a range that simultaneously meets the upper limit power value and the safe operating power range, thus obtaining the corrected power adjustment command of the over-limit equipment; The corrected power regulation command is encapsulated and merged with the refined power adjustment command of the device that has not been marked as exceeding the limit, forming the target coordination command for the large-scale user-side energy storage cluster.

10. A method for real-time coordinated control of large-scale user-side distributed energy storage clusters, based on the system described in any one of claims 1 to 9, characterized in that the steps include... include: S1. Based on the real-time operating status data of distributed energy storage devices and the electrical quantity data of grid connection points, the distributed energy storage devices are dynamically grouped to obtain the energy storage subgroups of distributed energy storage devices. S2. Based on the grid dispatch requirements and the division results of the energy storage subgroup, conduct a consistency assessment of the total power demand of the energy storage subgroup, and determine the grid dispatch instructions for the energy storage subgroup based on the assessment results and the equipment parameters of the energy storage subgroup. S3. Generate a global coordinated control target for the energy storage subgroup based on the grid dispatch instructions and the local voltage-frequency over-limit events of the energy storage subgroup; S4. Based on the global collaborative control objective, distributed collaborative optimization is performed on the edge collaborative nodes of the energy storage subgroup to obtain refined power adjustment commands for the distributed energy storage devices. S5. Based on the maximum chargeable and dischargeable power of the distributed energy storage device, perform a safety verification between the fine-grained power adjustment command and the current real-time operating status of the distributed energy storage device to obtain the target coordination command for the large-scale user-side energy storage cluster.

Images