A distributed energy storage system state data evaluation and cluster optimization regulation system and method

Abstract

The application provides a distributed energy storage system state data evaluation and cluster optimization regulation system and method, relates to the technical field of distributed energy storage, and comprises three units of data acquisition, data evaluation and optimization regulation, and a distributed energy storage system state data evaluation and cluster optimization regulation method applying the system. The application sets each energy storage unit as a distributed control node, accurately collects multi-dimensional operation state data, subdivides node types, matches different state evaluation methods for different types of nodes, accurately calculates SOC, SOH and various real-time available powers, formulates a differentiated power distribution strategy according to the accurate parameters, adjusts the power according to the actual capacity of each unit, improves the state evaluation accuracy and regulation pertinence of the energy storage cluster, avoids unreasonable scheduling of the aging units, prolongs the service life of the units, optimizes the power distribution efficiency of the cluster, and guarantees the safe, stable and efficient operation of the distributed energy storage system as a whole.

Description

Technical Field

[0001] This invention relates to the field of distributed energy storage technology, and more specifically, to a system and method for evaluating state data and optimizing the control of a distributed energy storage system cluster. Background Technology

[0002] With the rapid development of the new energy industry, distributed energy storage systems, as a core component of energy consumption and grid regulation, are seeing continuous expansion in application scale. The number of units within energy storage clusters is increasing, and their operating characteristics vary significantly. Simultaneously, the power system's requirements for the dispatch response speed, operational stability, and energy utilization efficiency of energy storage systems are constantly rising, making accurate status data assessment the foundation for cluster optimization and control. The health status, capacity specifications, and response characteristics of distributed energy storage units differ significantly, necessitating appropriate assessment and control methods to ensure the coordinated and efficient operation of energy storage clusters and the power system. This also places higher demands on the professionalism and precision of energy storage system status assessment and cluster control technologies.

[0003] Traditional distributed energy storage system cluster control methods often employ uniform state assessment standards and power allocation strategies, failing to differentiate node types based on the actual operating status, capacity, and response characteristics of energy storage units. The selection of assessment parameters is relatively simplistic, making it difficult to accurately calculate the SOC, SOH, and actual available power of each unit. The state assessment results lack specificity and accuracy. This type of control method easily leads to wasted capacity of high-performance units, while unreasonable overloading of aging units reduces the overall power dispatch efficiency of the energy storage cluster, accelerates the performance degradation of aging units, and may even cause operational failures, affecting the overall safety and stability of the energy storage system.

[0004] Therefore, it is necessary to design a distributed energy storage system state data assessment and cluster optimization control system and method to solve the problems of the lack of specificity and low calculation accuracy of the existing energy storage cluster state assessment, and the failure to differentiate the power allocation according to the actual operating capacity of each energy storage unit, resulting in low cluster control efficiency and unreasonable scheduling of aging units. Summary of the Invention

[0005] In view of this, the present invention proposes a distributed energy storage system state data assessment and cluster optimization control system and method, which aims to solve the problems of the lack of specificity and low calculation accuracy of existing energy storage cluster state assessment, and the failure to differentiate power allocation based on the actual operating capacity of each energy storage unit, resulting in low cluster control efficiency and unreasonable scheduling of aging units.

[0006] In one aspect, this invention proposes a system and method for evaluating state data and optimizing cluster control of distributed energy storage systems, comprising: The data acquisition unit is used to collect the operating status data of each control node in real time, with each energy storage unit as an independent local distributed control node. The operating status data includes response time, rated capacity, internal resistance, number of cycles, remaining power, terminal voltage, charging and discharging current and operating temperature, and determines the node type of each control node based on the operating status data. The data evaluation unit is used to assign different status data evaluation methods to each control node based on the node type, and to collect and calculate the precise status parameters of each control node in real time based on the status data evaluation methods. The precise status parameters include SOC, SOH and real-time available power. The real-time available power includes maximum available power, continuous available output power and limited available power. Each control node uploads the locally calculated precise status parameters to the distributed collaborative controller. An optimized control unit is used by the distributed collaborative controller to formulate differentiated power allocation strategies for each control node in the cluster based on the accurate status parameters uploaded by each control node, and to match the corresponding scheduling power according to the actual operating capacity of each control node.

[0007] Furthermore, when determining the node type of each control node based on the aforementioned operational status data, the process includes: When the response time is less than or equal to a preset millisecond-level response threshold, the rated capacity is less than or equal to a preset small capacity threshold, the internal resistance is less than or equal to a preset healthy internal resistance threshold, and the number of cycles is less than or equal to a preset healthy cycle number threshold, then the node type of the current control node is a distributed power-type healthy small capacity control node. When the rated capacity is greater than the preset small capacity threshold, the response time is greater than the preset millisecond response threshold, and the number of cycles is within the preset medium health cycle number range, then the node type of the current control node is a distributed energy type medium capacity medium health control node.

[0008] Furthermore, when determining the node type of each control node based on the aforementioned operational status data, the method also includes: When the internal resistance is greater than the preset healthy internal resistance threshold, the number of cycles is greater than the preset healthy cycle number threshold, and the remaining power is less than the preset remaining power threshold, then the node type of the current control node is a distributed decay type old control node.

[0009] Furthermore, when assigning different state data evaluation methods to each control node based on the node type, and when collecting and calculating the precise state parameters of each control node in real time based on the state data evaluation methods, the process includes: When the node type is a distributed power-type healthy small-capacity control node, a local state data evaluation method emphasizing fast response is adopted, and the current SOC, SOH and maximum available power of the current control node are obtained through local distributed calculation using the terminal voltage, the charging and discharging current and the operating temperature.

[0010] Furthermore, when assigning different state data evaluation methods to each control node based on the node type, and when collecting and calculating the precise state parameters of each control node in real time based on the state data evaluation methods, the method further includes: When the node type is a distributed energy type medium capacity medium health control node, a local energy-based state data evaluation method is adopted, and the SOC, SOH and continuous available output power of the current control node are obtained by local distributed calculation using the cumulative charge and discharge amount, the remaining power, the actual available capacity and the operating temperature.

[0011] Furthermore, when assigning different state data evaluation methods to each control node based on the node type, and when collecting and calculating the precise state parameters of each control node in real time based on the state data evaluation methods, the method further includes: When the node type is a distributed decay type old control node, a local state data evaluation method focusing on safety limitations is adopted, and the SOC, SOH and the limited available power of the current control node are obtained by local distributed calculation using internal resistance change, voltage deviation and the remaining power.

[0012] Furthermore, based on the precise status parameters uploaded by each control node, a differentiated power allocation strategy is formulated for each control node in the cluster. When matching the corresponding scheduling power according to the actual operating capacity of each control node, this includes: When the SOC is in a preset normal operating range, the SOH is greater than or equal to a preset health threshold, and the real-time available power is greater than a first power threshold, then a scheduling power greater than or equal to a first proportion of the rated power is allocated to the current node through distributed control instructions.

[0013] Furthermore, when formulating differentiated power allocation strategies for each control node in the cluster based on the precise status parameters uploaded by each control node, and matching the corresponding scheduling power according to the actual operating capacity of each control node, the process also includes: When the SOC is in a preset intermediate operating range and the SOH is in a preset medium health range, and the real-time available power is less than a first power threshold and greater than a second power threshold, then a scheduling power less than a first proportion of the rated power and greater than a second proportion of the rated power is allocated to the current node through distributed control instructions.

[0014] Furthermore, when formulating differentiated power allocation strategies for each control node in the cluster based on the precise status parameters uploaded by each control node, and matching the corresponding scheduling power according to the actual operating capacity of each control node, the process also includes: When the SOC is lower than the preset lower threshold or higher than the preset upper threshold, and the SOH is less than the preset health threshold and the real-time available power is less than or equal to the second power threshold, then the current node is allocated scheduling power less than or equal to the second proportion of the rated power through distributed control instructions.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: the distributed energy storage system status data assessment and cluster optimization control system of the present invention sets each energy storage unit as a distributed control node, accurately collects multi-dimensional operating status data and subdivides node types, matches different status assessment methods for different types of nodes, can accurately calculate SOC, SOH and various real-time available power, and then formulate differentiated power allocation strategies based on accurate status parameters to match the actual operating capacity of each energy storage unit for power scheduling. This not only improves the accuracy of energy storage cluster status assessment and the targeting of control, but also avoids unreasonable scheduling of aging units, effectively extends the service life of energy storage units, and optimizes cluster power allocation efficiency, ensuring the overall safe, stable and efficient operation of the distributed energy storage system.

[0016] On the other hand, this invention proposes a method for evaluating state data and optimizing cluster control of distributed energy storage systems, including: Each energy storage unit is treated as an independent local distributed control node. The operating status data of each control node is collected in real time. The operating status data includes response time, rated capacity, internal resistance, number of cycles, remaining power, terminal voltage, charging and discharging current and operating temperature. The node type of each control node is determined based on the operating status data. Based on the node type, different status data evaluation methods are assigned to each control node, and the precise status parameters of each control node are collected and calculated in real time based on the status data evaluation methods. The precise status parameters include SOC, SOH and real-time available power. The real-time available power includes maximum available power, continuous available output power and limited available power. Each control node uploads the locally calculated precise status parameters to the distributed collaborative controller. The distributed collaborative controller formulates differentiated power allocation strategies for each control node in the cluster based on the accurate status parameters uploaded by each control node, and matches the corresponding scheduling power according to the actual operating capacity of each control node.

[0017] It is understandable that the above-mentioned method and system for evaluating state data and optimizing cluster control of distributed energy storage systems have the same beneficial effects, and will not be elaborated further here. Attached Figure Description

[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a functional block diagram of a distributed energy storage system state data evaluation and cluster optimization control system provided in an embodiment of the present invention; Figure 2 A flowchart illustrating the distributed energy storage system status data assessment and cluster optimization control method provided in this embodiment of the invention. Detailed Implementation

[0019] Exemplary embodiments of the present invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present invention and to fully convey the scope of the invention to those skilled in the art. It should be noted that, without conflict, the embodiments and features described herein can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] Reference Figure 1 As shown in some embodiments of this application, a distributed energy storage system state data assessment and cluster optimization control system includes: The data acquisition unit is used to collect the operating status data of each control node in real time, with each energy storage unit as an independent local distributed control node. The operating status data includes response time, rated capacity, internal resistance, number of cycles, remaining power, terminal voltage, charging and discharging current and operating temperature, and determines the node type of each control node based on the operating status data. The data evaluation unit is used to assign different status data evaluation methods to each control node based on the node type, and to collect and calculate the precise status parameters of each control node in real time based on the status data evaluation methods. The precise status parameters include SOC, SOH and real-time available power. The real-time available power includes maximum available power, continuous available output power and limited available power. Each control node uploads the locally calculated precise status parameters to the distributed collaborative controller. An optimized control unit is used by the distributed collaborative controller to formulate differentiated power allocation strategies for each control node in the cluster based on the accurate status parameters uploaded by each control node, and to match the corresponding scheduling power according to the actual operating capacity of each control node.

[0021] Specifically, in this embodiment, the response time is the time interval from when the local distributed control node receives the control command issued by the distributed collaborative controller to when it completes the corresponding power adjustment execution action, used to characterize the command response rate of the energy storage unit, in milliseconds; the rated capacity is the nominal energy storage capacity calibrated at the time of manufacture of the energy storage unit, a fixed parameter reflecting the theoretical energy storage capacity of the energy storage unit; the internal resistance is the equivalent DC impedance inside the energy storage unit, used to reflect the internal impedance characteristics and the degree of charging and discharging loss of the energy storage unit; the cycle count is the total number of complete charge and discharge cycles completed by the energy storage unit since its commissioning, a core indicator for quantifying the aging degree of the energy storage unit; and the remaining power is the amount of remaining charge that the energy storage unit can actually output at the current moment. The parameters are dynamic parameters that reflect the current energy storage status of the energy storage unit in real time; the terminal voltage is the real-time voltage value at the positive and negative output terminals of the energy storage unit, which is the basic voltage parameter characterizing the real-time electrical status of the energy storage unit; the charging and discharging current is the real-time operating current of the energy storage unit under charging or discharging conditions, wherein the current is positive under charging conditions and negative under discharging conditions, which is used to reflect the real-time charging and discharging intensity of the energy storage unit; the operating temperature is the real-time operating temperature of the energy storage unit cell or module body, which is a key environmental parameter affecting the charging and discharging performance, aging rate and operational safety of the energy storage unit; in this embodiment, the response time is obtained by the high-precision timing chip built into the local distributed control node, which collects the power adjustment data sent by the distributed collaborative controller. The trigger time of the section command and the execution time of the corresponding power adjustment action of the energy storage unit are calculated by subtracting the timestamps of the two times to obtain a millisecond-level response time value; the rated capacity is obtained by reading the factory calibration parameters stored in the battery management system of the energy storage unit through the local distributed control node, or by calling the nominal capacity configuration data stored locally to directly obtain the rated capacity parameter of the energy storage unit; the internal resistance is obtained by the local distributed control node's internal resistance detection module outputting a low-frequency, small-amplitude detection voltage signal to the energy storage unit, simultaneously collecting the feedback current and corresponding voltage drop values ​​of the energy storage unit, and performing calculations based on Ohm's law to obtain the real-time equivalent internal resistance of the energy storage unit; the number of cycles is obtained by the local distributed control node... The cycle statistics module of the control node monitors the switching of the energy storage unit's charging and discharging state in real time. Each time a complete charge-discharge cycle from full charge to full discharge is completed, an accumulation count is performed to continuously accumulate the total number of cycles of the energy storage unit. The remaining power is obtained by real-time acquisition of terminal voltage and charging and discharging current through the local distributed control node. The change in power is calculated by time integration of the charging and discharging current using the ampere-hour integration method. Then, the error is corrected by the terminal voltage lookup table calibration method, and finally the real-time remaining power of the energy storage unit is obtained. The terminal voltage is obtained by connecting the high-precision voltage acquisition sensor on the local distributed control node in parallel to the positive and negative output terminals of the energy storage unit. After acquiring the analog voltage signal, it is amplified, filtered, and converted from analog to digital to obtain the accurate real-time terminal voltage value.The charging and discharging current is connected in series with the main charging and discharging circuit of the energy storage unit via a Hall current sensor mounted on the local distributed control node. After acquiring the real-time analog current signal, it undergoes signal conditioning, analog-to-digital conversion, and current direction determination to distinguish between positive charging current and negative discharging current, ultimately obtaining the real-time charging and discharging current value. The operating temperature is achieved by an NTC temperature sensor mounted on the local distributed control node, which is closely attached to the surface of the energy storage unit's cells or modules. After acquiring the temperature analog signal, it undergoes signal conditioning, analog-to-digital conversion, and temperature calibration to convert it into a real-time operating temperature value in degrees Celsius.

[0022] Understandably, by setting each energy storage unit as a distributed control node, accurately collecting multi-dimensional operational status data and subdividing node types, and matching differentiated status assessment methods to different types of nodes, it is possible to accurately calculate SOC, SOH, and various real-time available power. Then, based on the accurate status parameters, differentiated power allocation strategies can be formulated to match the actual operating capacity of each energy storage unit for power scheduling. This not only improves the accuracy of energy storage cluster status assessment and the targeting of regulation, but also avoids unreasonable scheduling of aging units, effectively extending the service life of energy storage units, while optimizing cluster power allocation efficiency and ensuring the overall safe, stable, and efficient operation of the distributed energy storage system.

[0023] In some embodiments of this application, determining the node type of each control node based on the operational status data includes: When the response time is less than or equal to a preset millisecond-level response threshold, the rated capacity is less than or equal to a preset small capacity threshold, the internal resistance is less than or equal to a preset healthy internal resistance threshold, and the number of cycles is less than or equal to a preset healthy cycle number threshold, then the node type of the current control node is a distributed power-type healthy small capacity control node. When the rated capacity is greater than the preset small capacity threshold, the response time is greater than the preset millisecond response threshold, and the number of cycles is within the preset medium health cycle number range, then the node type of the current control node is a distributed energy type medium capacity medium health control node.

[0024] Specifically, in this embodiment, the preset millisecond-level response threshold is a millisecond-level critical value pre-set by the system to determine the response rate of the energy storage unit. It is a core judgment parameter for distinguishing whether an energy storage unit has fast response characteristics and directly determines the screening criteria for power-type energy storage units. Its value setting is in line with the system's regulation and response requirements, and is used to define the upper limit of the time from receiving an instruction to executing an action by the energy storage unit. The preset small capacity threshold is a capacity critical value pre-calibrated by the system for the energy storage unit. It is a quantitative judgment basis for classifying the capacity level of energy storage units and clarifying the distinction between power-type small-capacity units and energy-type medium-capacity units. Its value is determined in combination with the energy storage unit product specifications and the capacity matching requirements of the system cluster regulation. The preset healthy internal resistance threshold is a critical value for the internal impedance of the energy storage unit pre-set by the system. It is a judgment standard for characterizing that the internal impedance of the energy storage unit is in a state of no obvious aging and can operate normally and efficiently. It is used to distinguish whether the internal resistance of the energy storage unit exceeds the healthy operating range and avoid increased charging and discharging losses due to excessive internal resistance. The preset healthy cycle number threshold is a critical value for the number of charge and discharge cycles of the energy storage unit pre-set by the system. It is used to define whether the energy storage unit is in a healthy service cycle and has not experienced any problems. The core parameter for determining obvious cyclic aging is referenced to the energy storage unit's cycle life design index, used to determine whether the energy storage unit is in the initial healthy operation stage. The preset medium healthy cycle number range is a numerical range pre-set by the system between the preset healthy cycle number threshold and the aging cycle number threshold. It is the basis for determining whether the energy storage unit is in a medium healthy state, not completely aged but with slight performance degradation, used to distinguish healthy energy storage units from degraded energy storage units, and to provide a quantitative standard for screening medium-capacity energy-type units. In this embodiment, the preset millisecond-level response threshold is combined with the overall control response rate requirement of the energy storage system (≤60ms) and the design response index of the power-type energy storage unit (≤50ms). Referring to the relevant provisions on response time in GB / T36547-2018 "Electrochemical Energy Storage System Energy Storage Unit Performance Test Method", 10 groups of power-type energy storage units of the same specification (rated capacity 5kWh, rated power 10kW) were selected for simulated command response test. The high-precision timing chip built into the local distributed control node recorded the 0.0 ...The trigger time (T1) of the 5x rated power adjustment command and the execution time (T2) of the completed power adjustment action were calculated. The response time ΔT = T2 - T1 for each unit was calculated. Each unit was tested 30 times, and a total of 300 sets of response time data were collected. Outliers exceeding the 3σ range (such as outliers ≤10ms or ≥80ms) were removed using the Grubbs criterion. Statistical analysis was performed on the remaining valid data, and the value corresponding to the 95th percentile was taken as the critical value that can meet the rapid adjustment requirements of the system (such as 50ms). Then, 100 rapid adjustment scenarios were simulated through system joint debugging to verify that this critical value can accurately distinguish between fast-response (ΔT≤50ms) and non-fast-response (ΔT>50ms) units. The threshold is pre-written into the Flash non-volatile storage module of each local distributed control node, with an independent read address set to ensure stable storage and direct access by the node type determination module. The preset small capacity threshold is based on the energy storage industry's energy storage unit capacity grading standards (small capacity ≤ 5kWh, medium capacity 5-20kWh). Combined with the design division and functional positioning of power-type (focusing on fast response) and energy-type (focusing on capacity storage) energy storage units in this system, the capacity range (2-20kWh) of 15 types of energy storage units compatible with this system is statistically analyzed. The capacity and power density data of various units are fitted and analyzed using Excel software, combined with the capacity matching requirements of the system cluster control (power-type...). Based on the unit capacity ≤ 5kWh (allowing for flexible scheduling) and the power output capability of each unit (power density ≥ 2kW / kWh for small-capacity units), a capacity threshold (e.g., 5kWh) is determined to effectively distinguish between the two types of units. After system integration and testing with 5 groups of power-type (2-5kWh) and 5 groups of energy-type (6-20kWh) units, and verification that the threshold can stably achieve capacity classification with 100% stability, this threshold is pre-stored in the EEPROM parameter configuration module of the distributed collaborative controller and synchronized to each local control node via the CAN bus to generate a threshold comparison index table, supporting direct comparison during subsequent node type determination. The preset healthy internal resistance threshold is based on the nominal internal resistance of the energy storage unit at the time of manufacture (e.g., 20mΩ). Based on the benchmark and referring to the industry standard for the allowable fluctuation range of the healthy internal resistance of energy storage units (≤10%), eight sets of brand-new, non-aging energy storage units of the same specifications were selected. The internal resistance detection module of the local distributed control node outputs a low-frequency, small-amplitude detection voltage signal of 1kHz and 5V to each unit, and the feedback current signal of each unit is collected simultaneously. Based on Ohm's law R=U / I, the initial healthy internal resistance of each set of units is calculated, and the mean of the eight sets of data is calculated (e.g., 19.8mΩ). Combining the internal resistance decay law of energy storage units during long-term operation (decay ≤2% per year), a reasonable critical value (e.g., 21.8mΩ) is set to accurately distinguish between healthy and slightly aged units. The threshold accuracy is repeatedly calibrated by the internal resistance detection module (each calibration deviation ≤±0).1mΩ), select 10 slightly aged units (internal resistance 22-25mΩ) for verification to ensure that the judgment deviation caused by internal resistance detection error can be effectively avoided. After that, the threshold is pre-configured to the threshold storage unit of the data acquisition unit and linked with the real-time internal resistance acquisition module. Set the threshold comparison trigger mechanism for node type determination; The preset healthy cycle number threshold is based on the cycle life index (e.g., 5000 times) of the energy storage unit at the factory, combined with the design requirements of the healthy service cycle of the energy storage unit (capacity decay rate ≤10%), and referring to the relevant test data of energy storage unit cycle aging in the industry, select 6 groups of energy storage units of the same specification for accelerated aging test (ambient temperature 45℃, charge / discharge rate 1C). After every 100 charge / discharge cycles, the actual capacity of each unit is measured by the capacity detection module. Record the maximum number of cycles for which each unit does not have significant performance degradation (capacity decay rate ≤10%). Use Origin software to fit the relationship curve between the number of cycles and the capacity decay rate to determine the critical value (e.g., 2000 times). After the critical value is preset and entered into the system, select 3 groups of units for long-term operation verification (actual After confirming the feasibility of the threshold (if the capacity decay rate is still ≤10% after 2000 cycles), the threshold is pre-stored in the cycle statistics module of the data evaluation unit, and cycle count comparison logic is set to compare and judge with the real-time collected cycle count data; the preset medium healthy cycle count range takes the preset healthy cycle count threshold (2000 cycles) as the lower limit, and the cycle count threshold (e.g., 5000 cycles) that indicates significant performance degradation (capacity decay rate reaches 20%) in the accelerated aging test of the energy storage unit as the upper limit, referring to the actual values ​​of 10 groups of similar energy storage units. Using aging data (capacity decay rate of 10%-20% after 2000-5000 cycles), a range verification method was employed to select 5 groups of units within this range. Performance testing confirmed their health status as moderate (normal operation possible but with a slight decrease in power output). After system integration and connection of 3 groups of units within this range, verifying that this range accurately determines the moderate health status of energy storage units, this range (2001 to 5000 cycles) was pre-stored in the distributed collaborative controller and each local control node. A range comparison algorithm was then generated to ensure the accuracy of node type determination.

[0025] Specifically, in this embodiment, the preset millisecond-level response threshold is set to 50 milliseconds, the preset small capacity threshold is set to 5 kilowatts, the preset healthy internal resistance threshold is set to 20 milliohms, the preset healthy cycle count threshold is set to 2000 times, and the preset medium healthy cycle count range is set to 2001 to 5000 times. The system collects the operating status data of each distributed control node in real time through the data acquisition unit. Among them, the response time of a certain distributed control node is collected as 30 milliseconds, the rated capacity is collected as 3 kilowatts, the internal resistance is collected as 15 milliohms, and the cycle count is collected as 1500 times. The data acquisition unit compares the response time of this node with the preset millisecond-level response threshold of 50 milliseconds, the rated capacity with the preset small capacity threshold of 5 kilowatts, the internal resistance with the preset healthy internal resistance threshold of 20 milliohms, and the cycle count with the preset healthy cycle count threshold of 2000 times. Because the response time of this node is 30 milliseconds... The node with a response time of 70 milliseconds ≤ 50 milliseconds, rated capacity of 3 kWh ≤ 5 kWh, internal resistance of 15 milliohms ≤ 20 milliohms, and cycle count of 1500 ≤ 2000 is determined to be a distributed power type healthy small capacity control node. The response time of another distributed control node is 70 milliseconds, rated capacity is 10 kWh, internal resistance is 18 milliohms, and cycle count is 3500. The data acquisition unit compares the rated capacity of this node with the preset small capacity threshold of 5 kWh, the response time with the preset millisecond response threshold of 50 milliseconds, and the cycle count with the preset medium health cycle count range of 2001 to 5000. Since the rated capacity of this node (10 kWh > 5 kWh), the response time (70 milliseconds > 50 milliseconds), and the cycle count (3500) are within the range of 2001 to 5000, this node is determined to be a distributed energy type medium capacity medium health control node.

[0026] Understandably, the quantitative judgment criteria for two types of core nodes—distributed power-type healthy small-capacity control nodes and distributed energy-type medium-capacity medium-health control nodes—have been clearly defined, enabling accurate initial classification of energy storage units. The quantitative node judgment criteria, based on multi-dimensional operating status data such as response time, rated capacity, internal resistance, and cycle count, clearly define energy storage units with different capacity specifications, health states, and response characteristics. This provides a precise classification basis for matching customized status data assessment methods to different nodes, avoiding the problem of homogenization of assessment methods at the node judgment level and significantly improving the pertinence of status data assessment.

[0027] In some embodiments of this application, when determining the node type of each control node based on the operating status data, the method further includes: When the internal resistance is greater than the preset healthy internal resistance threshold, the number of cycles is greater than the preset healthy cycle number threshold, and the remaining power is less than the preset remaining power threshold, then the node type of the current control node is a distributed decay type old control node.

[0028] Specifically, in this embodiment, the preset remaining power threshold is a critical value for the remaining power of the energy storage unit pre-set by the system. It is an auxiliary quantitative parameter characterizing whether the current remaining power of the energy storage unit is lower than the minimum standard for safe operation and determining the node type in conjunction with the aging status of the energy storage unit. Its value is determined by combining the rated capacity of the energy storage unit, the low power operation protection requirements, and the power decay characteristics of the aging unit. It is used to compare with the real-time collected remaining power data, and together with the preset healthy internal resistance threshold and the preset healthy cycle number threshold, jointly define the distributed decaying aging control nodes, avoid the energy storage unit from aging faster due to long-term low power operation or the emergence of operational safety hazards, and at the same time provide a clear electrical value for the accurate determination of the node type. The remaining power threshold is determined based on the rated capacity of the energy storage unit (selecting a 10kWh energy storage unit), the industry standard for low power protection of energy storage units (the remaining power at low power should not be less than 15% of the rated capacity), and the power decay law of aging units (the low power safety boundary of aging units is 5% lower than that of healthy units). Referring to Article 5.3 of GB / T36547-2018 "Electrochemical Energy Storage System Energy Storage Unit Performance Test Method" regarding the specific provisions on remaining power protection, eight groups of energy storage units of the same specification that have shown obvious aging were selected (the aging determination criteria are: internal resistance > 21.8mΩ, cycle count > 2000 times, compared with the preset healthy internal resistance threshold). (Based on a preset healthy cycle number threshold), the CT-4008 battery testing system was used to conduct low-charge discharge tests on the 8 groups of aged cells. The discharge rate was set to 0.2C. Real-time data on the terminal voltage, remaining capacity, and operating temperature of each group of cells were collected. When the cell terminal voltage dropped to 3.0V (the lower limit voltage for safe discharge of the energy storage unit), the operating temperature stabilized at 25±2℃, and the discharge current fluctuation was ≤0.1A, the remaining capacity at this time was recorded as the minimum remaining capacity for safe operation of the cell. The test was repeated 6 times for each group of cells, and a total of 48 sets of minimum remaining capacity data were collected. The Grubbs criterion was used to remove outliers exceeding the 3σ range (such as outliers <1.5kWh or >2.5kWh). The remaining valid data were arithmetically averaged to obtain a mean of 2.0 kWh. Considering the system's low-power protection strategy (reserving 5% of rated capacity as emergency power) and the stability requirements of cluster regulation, this mean was calibrated by ±0.2 kWh. Three candidate thresholds of 1.8 kWh, 2.0 kWh, and 2.2 kWh were tested. Fifty node judgment scenarios were simulated through system integration testing, and the misjudgment rate of each candidate threshold was statistically analyzed. The system was ensured that the misjudgment rate was <1%, accurately distinguishing between low-power aging units (remaining power < threshold) and normal units (remaining power ≥ threshold), and that a threshold setting that was too high would not cause normal units to be misjudged as aging units, while a threshold setting that was too low would prevent aging units from being identified in a timely manner. Therefore, the value of 2.0 kWh was adjusted accordingly.A preset remaining power threshold of 0 kWh (corresponding to 20% of the 10 kWh rated capacity) is pre-written into the Flash threshold storage module of each local distributed control node and the parameter configuration unit of the distributed collaborative controller. An independent threshold reading interface is set up, and threshold comparison logic is generated to ensure that this threshold can be directly called during node type determination and accurately compared with the real-time collected remaining power data, guaranteeing the accuracy of the determination result.

[0029] Specifically, in this embodiment, the preset healthy internal resistance threshold is set to 20 milliohms, the preset healthy cycle count threshold is set to 2000 times, and the preset remaining power threshold is set to 20% of the rated capacity (if the rated capacity of the corresponding energy storage unit is 10 kWh, then the preset remaining power threshold is specifically 2 kWh). The system collects the operating status data of a certain distributed control node in real time, such as internal resistance, cycle count, and remaining power, through the data acquisition unit. The real-time acquired value of the node's internal resistance is 25 milliohms, the real-time acquired value of the cycle count is 2800 times, and the real-time acquired value of the remaining power is... The value is 1.8 kWh (corresponding to 18% of the rated capacity of 10 kWh). The data acquisition unit compares the node's internal resistance with the preset healthy internal resistance threshold of 20 milliohms, the number of cycles with the preset healthy cycle count threshold of 2000, and the remaining power with the preset remaining power threshold of 2 kWh. Since the node's internal resistance of 25 milliohms > 20 milliohms, the number of cycles of 2800 > 2000, and the remaining power of 1.8 kWh < 2 kWh, all the judgment conditions are met. Therefore, the node is determined to be a distributed attenuation type old control node.

[0030] Understandably, the addition of quantitative judgment conditions for distributed decaying aging control nodes improves the node type judgment and classification system for the entire life cycle of energy storage units, achieving full coverage identification of energy storage units in three different operating states: healthy, moderately healthy, and aging. This allows energy storage units in the aging stage to be accurately and individually defined, making the system's node classification more closely match the actual operating state of energy storage units. It also provides a clear basis for matching aging units with dedicated state assessment methods that emphasize safety limitations. From the classification level, it avoids the problem of using conventional assessment methods and making unreasonable scheduling for aging units in advance, providing classification-level protection for the safe operation of aging units.

[0031] In some embodiments of this application, when assigning different state data evaluation methods to each control node based on the node type, and when collecting and calculating the precise state parameters of each control node in real time based on the state data evaluation methods, the process includes: When the node type is a distributed power-type healthy small-capacity control node, a local state data evaluation method emphasizing fast response is adopted, and the current SOC, SOH and maximum available power of the current control node are obtained through local distributed calculation using the terminal voltage, the charging and discharging current and the operating temperature.

[0032] Specifically, in this embodiment, when performing local distributed calculations on the distributed power-type healthy small-capacity control node, the node's terminal voltage is collected in real time as 3.2V, charging / discharging current as 10A, and operating temperature as 25℃. The reference operating temperature is set to 25℃, the temperature compensation coefficient is 1.0, the rated capacity is 5Ah, the rated internal resistance is 20mΩ, and the rated cycle life is 5000 cycles. First, the initial state of charge is obtained by substituting the real-time terminal voltage of 3.2V into the preset terminal voltage-SOC reference curve. Then, calibration calculations are performed using the temperature compensation coefficient of 1.0 corresponding to the operating temperature of 25℃, resulting in SOC = (initial voltage corresponding to state of charge × temperature compensation coefficient) = 80%. Subsequently, based on the real-time internal resistance... With a rated internal resistance of 15mΩ, a rated internal resistance of 20mΩ, 1500 operating cycles, and a rated cycle life of 5000 cycles, the SOH is calculated as (rated internal resistance ÷ real-time internal resistance) × (1 - number of operating cycles ÷ rated cycle life) to obtain SOH = (20mΩ ÷ 15mΩ) × (1 - 1500 ÷ 5000) = 93%. Finally, the real-time terminal voltage of 3.2V is multiplied by the maximum allowable charge / discharge current of 15A after temperature compensation. The maximum available power is calculated as 3.2V × 15A = 48W using the formula: maximum available power = real-time terminal voltage × maximum allowable charge / discharge current after temperature compensation. This completes the local distributed calculation of SOC, SOH, and maximum available power for this control node.

[0033] Understandably, a local state data evaluation method emphasizing rapid response was customized for distributed power-type healthy small-capacity control nodes, and suitable calculation parameters were clearly defined. This method aligns with the core characteristics of this type of node: rapid response, healthy operation, and small-capacity power. Adaptive parameters such as terminal voltage, charging and discharging current, and operating temperature were selected for local distributed calculation. This ensures the accuracy of the SOC, SOH, and maximum available power calculation results, while also fitting the rapid response function positioning of this type of node. At the same time, the local distributed calculation method significantly improves the efficiency of state parameter acquisition, enabling accurate and rapid output of the core state parameters of this type of node, perfectly adapting to the system's rapid regulation requirements for power-type nodes.

[0034] In some embodiments of this application, when assigning different state data evaluation methods to each control node based on the node type, and when collecting and calculating the precise state parameters of each control node in real time based on the state data evaluation methods, the method further includes: When the node type is a distributed energy type medium capacity medium health control node, a local energy-based state data evaluation method is adopted, and the SOC, SOH and continuous available output power of the current control node are obtained by local distributed calculation using the cumulative charge and discharge amount, the remaining power, the actual available capacity and the operating temperature.

[0035] Specifically, in this embodiment, when performing local distributed accounting for the medium-capacity, medium-health control node in a distributed energy system, the node's cumulative charge / discharge is collected in real time as 1200 kWh, remaining capacity is 8 kWh, actual usable capacity is 10 kWh, and operating temperature is 28°C. The node's rated capacity is set to 12 kWh, rated power to 5 kW, with a cumulative charge / discharge correction factor of 0.99, an operating temperature correction factor of 0.98, and an operating temperature power factor of 0.98. First, using the formula SOC = Remaining Capacity ÷ Actual Usable Capacity × 100%, the values ​​are substituted to obtain SOC = 8 kWh ÷ 10 kW. Wh × 100% = 80%. Then, using the formula SOH = (Actual available capacity ÷ Rated capacity) × Cumulative charge / discharge correction factor × Operating temperature correction factor, we can calculate SOH = (10kWh ÷ 12kWh) × 0.99 × 0.98 ≈ 81.7%. Finally, using the formula Continuous available output power = (Remaining power ÷ Actual available capacity) × Rated power × Operating temperature power factor, we can calculate Continuous available output power = 80% × 5kW × 0.98 = 3.92kW. This completes the local distributed calculation of SOC, SOH, and continuous available output power for this control node.

[0036] Understandably, a local energy metering-focused status data assessment method was customized for distributed energy-type medium-capacity, medium-health control nodes. Suitable calculation parameters were selected, aligning with the node's operational characteristics of emphasizing energy storage, medium capacity, and medium health. Parameters highly correlated with energy metering, such as cumulative charge / discharge and actual available capacity, were incorporated into the calculation system to accurately calculate SOC, SOH, and continuously available output power. The calculation process considers both the core energy storage function of these nodes and their medium-health status, ensuring the assessment results better reflect the actual operational capabilities of these nodes and providing precise parameter support for subsequent reasonable power scheduling.

[0037] In some embodiments of this application, when assigning different state data evaluation methods to each control node based on the node type, and when collecting and calculating the precise state parameters of each control node in real time based on the state data evaluation methods, the method further includes: When the node type is a distributed decay type old control node, a local state data evaluation method focusing on safety limitations is adopted, and the SOC, SOH and the limited available power of the current control node are obtained by local distributed calculation using internal resistance change, voltage deviation and the remaining power.

[0038] Specifically, in this embodiment, when performing local distributed accounting for the distributed attenuation type aging control node, the internal resistance change of the node is collected in real time as 5 milliohms, the voltage deviation is 0.2V, and the remaining power is 1.8kWh. The rated internal resistance of the node is set to 20 milliohms, the rated terminal voltage is 3.2V, and the rated power is 3kW. First, the voltage deviation coefficient is calculated as: Voltage Deviation ÷ Rated Terminal Voltage = 0.2V ÷ 3.2V = 0.0625. Then, using the formula SOC = Remaining Power × (1 - Voltage Deviation Coefficient), the values ​​are substituted to obtain SOC = 1.8kWh × (1 - 0.0625) = 1.6875kWh (corresponding to the rated...). With a base SOC of 16.875% for a capacity of 10kWh, the formula SOH = rated internal resistance ÷ (rated internal resistance + internal resistance change) × 100% is used to calculate SOH = 20 milliohms ÷ (20 milliohms + 5 milliohms) × 100% = 80%. Finally, the internal resistance attenuation coefficient is set to 0.8 and the voltage deviation attenuation coefficient is set to 0.9. Using the formula for restricted available power = rated power × internal resistance attenuation coefficient × voltage deviation attenuation coefficient, the restricted available power is calculated to be 3kW × 0.8 × 0.9 = 2.16kW. This completes the local distributed calculation of SOC, SOH, and restricted available power for this control node.

[0039] Understandably, a local state data assessment method emphasizing safety constraints was customized for distributed decaying aging control nodes. This method clearly defines core calculation parameters such as internal resistance change, voltage deviation, and remaining power. Closely aligned with the operating characteristics of aging units, it conducts state calculations based on parameters that directly reflect the degree of aging and operational safety status of the units. It accurately obtains SOC, SOH, and limited available power. The calculation process consistently emphasizes the core principle of safety constraints, ensuring that the assessment results accurately reflect the actual state of the aging units while also demonstrating their safe operating boundaries. From a state assessment perspective, this provides accurate and reliable parameter basis for the subsequent safe scheduling of aging units, avoiding unreasonable scheduling problems caused by assessment biases.

[0040] In some embodiments of this application, when formulating differentiated power allocation strategies for each control node in the cluster based on the precise status parameters uploaded by each control node, and matching the corresponding scheduling power according to the actual operating capacity of each control node, the process includes: When the SOC is in a preset normal operating range, the SOH is greater than or equal to a preset health threshold, and the real-time available power is greater than a first power threshold, then a scheduling power greater than or equal to a first proportion of the rated power is allocated to the current node through distributed control instructions.

[0041] Specifically, in this embodiment, the preset normal operating range is a pre-set SOC (State of Charge) value range of the system, which is the optimal operating range characterizing the energy storage unit's state of charge as safe, stable, and capable of stable power output. It is used to determine whether the current power of the energy storage unit meets the basic conditions for main power dispatch, avoiding the impact of abnormal power on the operating performance and service life of the energy storage unit. The preset health threshold is a pre-set SOH (State of Health) critical value of the system, which is the core criterion for determining that the energy storage unit is in good health, has no obvious performance degradation, and can undertake main control tasks. It is used to distinguish between healthy energy storage units and aging energy storage units. The first power threshold is a pre-set real-time available power critical value of the system. The first proportion of rated power is a quantitative basis for determining whether an energy storage unit has sufficient power output capability and can be allocated a high proportion of dispatch power. It is used to screen energy storage nodes that can execute the main control command. The first proportion of rated power is the rated power allocation ratio preset by the system. It is the proportion standard for the distributed collaborative controller to allocate the main dispatch power to energy storage nodes that meet all healthy operating conditions. It is used to determine the share of main control power that such nodes can undertake. In this embodiment, the preset normal operating range is based on the safety operation specification of electrochemical energy storage unit (Article 4.2 of GB / T36547-2018 "Electrochemical Energy Storage System Energy Storage Unit Performance Test Method") and the charge and discharge operating characteristics of the energy storage unit (charge and discharge rate 0).Based on the optimal performance at 3C-1C and the overall system scheduling requirements (the main control node needs to have a stable output power ≥4kW), and considering the optimal state of charge (SOC) operating range of power-type energy storage units (5kWh / 10kW) and energy-type energy storage units (10kWh / 5kW), a system simulation model was built using PSCAD / EMTDC simulation software. This model simulates the power output stability, capacity decay rate, and operational safety of the two types of units within different SOC ranges (40%-100%) under three ambient temperatures: 25℃, 35℃, and 15℃. Power output data for each SOC range was collected. Fluctuation values ​​and capacity decay rate data were obtained. Simultaneously, six power-type and six energy-type units were selected for actual operating condition debugging. A charge-discharge cycle of 30 times was set, and the operating parameters of each unit under different SOCs were collected in real time. Origin software was used to fit and analyze the simulation and measured data. Ranges with power fluctuations >5% when SOC < 60% and capacity decay rates >1% when SOC > 90% were eliminated, determining 60%-90% as a reasonable normal operating SOC range. After 100 simulations of main scheduling scenarios in system integration, it was verified that the units within this range could stably output power without significant performance degradation. After reduction, the interval parameters are pre-written into the EEPROM parameter configuration module of the distributed collaborative controller to set the interval comparison logic and support subsequent SOC status determination. The preset health threshold is based on the factory health status calibration value of the energy storage unit (SOH=100% for brand new units) and the industry energy storage system health status determination standard (SOH≥90% is a healthy state). Twelve groups of energy storage units of the same specification with different aging levels are selected (2 groups of brand new units, 4 groups of slightly aged units, 4 groups of moderately aged units, and 2 groups of severely aged units). The CT-4008 battery performance testing system is used to measure the actual performance of each group of units. The actual capacity, internal resistance, and power output capability were considered. The SOH value of each unit was calculated (SOH = actual capacity / rated capacity × 100%). The power output stability of the unit corresponding to different SOH values ​​was statistically analyzed. When SOH ≥ 90%, the unit power output fluctuation was ≤ 3%, and the internal resistance showed no obvious abnormalities, allowing it to stably undertake the main control task. When SOH < 90%, the power output capability decreased significantly. The Grubbs criterion was used to eliminate abnormal test data, and the effective data was statistically calibrated to determine 90% as the critical SOH value for stable power output. Multiple precision calibrations were performed (calibration deviation ≤ ±0).After 50 actual operating condition verifications without error, the first power threshold is pre-stored in the Flash storage module of each local distributed control node and linked with the SOH calculation module. The first power threshold is combined with the rated power indicators of different types of energy storage units (rated power of 10kW for power-type units and 5kW for energy-type units) and the minimum main power requirement for system cluster control (minimum output power of single-node main control ≥ 4kW). Eight power-type and eight energy-type units are selected for power output testing. The operating temperature is set to 25℃ and SOC = 75% (midpoint of normal operating range). The maximum real-time available power of each unit is collected in real time. The minimum available power value that meets the main control requirement for each type of unit is calculated. The collected data is averaged, and outliers exceeding 3σ are removed, resulting in an average of 4.2kW. Combined with the system cluster control redundancy requirement (5% redundancy reserved), the average is calibrated to 4kW. After system integration and connection to 10 main nodes, it is verified that the threshold can accurately screen nodes with main control capability (real-time available power > 4kW) without omissions or misjudgments. This threshold is then pre-configured in the power judgment of the optimized control unit. The module sets a threshold comparison trigger mechanism. The first proportion of rated power is based on the system power allocation strategy (main nodes bear the main control load while taking into account the unit's operating life), the control task weight of energy storage nodes (main node weight ≥ 80%), and the actual power output capacity of various energy storage units (power-type units can reach a maximum output power of 1.1 times the rated power, and energy-type units can reach 1.0 times the rated power). A power allocation simulation model is built using MATLAB / Simulink to simulate the system scheduling stability, unit power loss, and lifespan impact under different allocation ratios (70%, 75%, 80%, 85%). Data on scheduling response time and unit power loss rate corresponding to each ratio are collected. Simultaneously, five groups of main nodes are selected for actual scheduling tests. The allocation ratio is adjusted multiple times, and the scheduling effect under each ratio is statistically analyzed. 80% is determined to be the optimal allocation ratio (at which the scheduling response is fastest, power loss is smallest, and the impact on unit lifespan is lowest). After verifying its feasibility through 100 scheduling simulations, this ratio is pre-set in the power allocation algorithm module of the distributed collaborative controller to generate the ratio allocation formula, ensuring that the main nodes can accurately obtain the corresponding proportion of scheduling power.

[0042] Specifically, in this embodiment, the preset normal operating range is set to SOC 60%~90%, the preset health threshold is set to SOH 90%, the first power threshold is set to 4kW, and the first proportion of rated power is set to 80%. The distributed collaborative controller receives precise status parameters uploaded by a distributed control node. The calculated SOC value of this node is 85%, the calculated SOH value is 95%, and the calculated real-time available power value is 5kW. The distributed collaborative controller compares the node's SOC with the preset normal operating range of 60%~90%, the SOH with the preset health threshold of 90%, and the real-time available power with the first power threshold of 4kW. The node's SOC is within the preset normal operating range, its SOH is greater than or equal to the preset health threshold, and its real-time available power is greater than the first power threshold, satisfying all the judgment conditions. Then, through distributed control commands, the distributed controller allocates scheduling power greater than or equal to 80% of its rated power to the node. The node's rated power is 5kW, and the corresponding allocated scheduling power is greater than or equal to 4kW.

[0043] Understandably, based on the precise status parameters uploaded by each node, a high-proportion power allocation strategy was formulated for nodes in high-quality operating states. Through quantitative judgment conditions of SOC, SOH, and real-time available power, it accurately identifies energy storage units that are in normal operating range, in excellent health, and have sufficient power output capacity, and allocates dispatch power greater than or equal to the first proportion of rated power to them. This allows high-performance energy storage units to fully undertake the main control load, maximizing their high-quality operating capabilities. This not only improves the overall power dispatch efficiency of the energy storage cluster, but also makes power allocation more scientific and targeted through quantitative judgment conditions, avoiding the waste of the capacity of high-quality nodes.

[0044] In some embodiments of this application, when formulating differentiated power allocation strategies for each control node in the cluster based on the precise status parameters uploaded by each control node, and matching the corresponding scheduling power according to the actual operating capacity of each control node, the method further includes: When the SOC is in a preset intermediate operating range and the SOH is in a preset medium health range, and the real-time available power is less than a first power threshold and greater than a second power threshold, then a scheduling power less than a first proportion of the rated power and greater than a second proportion of the rated power is allocated to the current node through distributed control instructions.

[0045] Specifically, in this embodiment, the preset intermediate operating range is a pre-set SOC (State of Charge) value range, which is between the preset normal operating range and the low power warning range. It characterizes the energy storage unit's state of charge as being at a moderate level, possessing stable but non-peak power support capability, and is the core power determination criterion for determining whether the energy storage unit is suitable for medium-intensity power dispatch tasks. The preset moderate health range is a pre-set SOH (State of Health) value range, which is between the preset health threshold and the attenuation-type unit determination threshold. It characterizes the energy storage unit as having slight performance degradation, a moderate health status, and being able to participate in regular power dispatch but not suitable for peak control tasks. It is the quantitative standard for distinguishing moderately healthy energy storage units from excellent and attenuated units. The second power threshold is a pre-set real-time available power threshold, which is lower than the first power threshold. It is the key numerical basis for determining whether the energy storage unit only has moderate power output capability and can only be allocated moderate-amplitude dispatch power. The second ratio of rated power is a pre-set value... The rated power allocation ratio, which is lower than the first ratio of rated power, is a fixed ratio standard for the distributed collaborative controller to allocate moderate-amplitude dispatch power to energy storage nodes in a moderate operating state or moderate health state. In this embodiment, the preset intermediate operating range is divided based on the full life cycle SOC operating characteristics of the electrochemical energy storage unit and the power allocation requirements of the system's three-level power dispatch. The upper limit is the preset normal operating range of 60% to 90%, and the lower limit is the low power warning range of 30%. Referring to the SOC operating range division specification in GB / T36547-2018 "Electrochemical Energy Storage System Energy Storage Unit Performance Test Method", the system model is built using PSCAD / EMTDC simulation software to simulate three typical ambient temperatures of 15℃, 25℃, and 35℃. Charge and discharge simulation tests are conducted on the energy storage unit in the 30% to 60% SOC range. The power output fluctuation value, capacity decay rate, and operating stability data of the unit in this range are collected in real time. At the same time, 6 groups of healthy energy storage units are selected for actual charge and discharge debugging, with 0.The 3C rate was tested for 30 cycles, and the corresponding operating parameters were recorded. Origin software was used to fit and analyze the simulation and measured data, verifying that the unit power output fluctuation within this range was ≤5%, there were no safety hazards, and it could stably match medium scheduling requirements. After determining 30%–60% as the preset intermediate operating range, the parameters of this range were pre-written into the EEPROM parameter configuration module of the distributed collaborative controller, configuring the range comparison and judgment logic. The preset medium health range was determined based on the energy storage unit's SOH decay law, the health status distribution characteristics under different cycle counts, combined with a preset health threshold of 90% and a decaying unit SOH critical value of 70%. Ten groups of energy storage units of the same specification in a slightly aging state were selected. The actual usable capacity, internal resistance, and power output capability of each group of cells were measured using the CT-4008 battery performance testing system. The SOH value for each group was calculated as SOH = Actual Usable Capacity / Rated Capacity × 100%. The scheduling adaptability of cells within different SOH ranges was statistically analyzed. It was confirmed that when the SOH is between 70% and 90%, the cells can stably perform medium-level scheduling tasks without significant operational faults. Abnormal test data was removed using the Grubbs criterion, and the range boundary calibration was completed. After verifying that this range can effectively distinguish between medium-health cells and healthy / degraded cells, the preset medium-health range of 70%–90% was pre-stored in the Flash storage module of each local distributed control node. The second power threshold is based on the first power... Using a power threshold of 4kW as a benchmark, eight energy storage units in a medium power and medium health state were selected for power output testing. The operating temperature was set at 25℃, SOC at 45% (median value of the intermediate operating range), and SOH at 80% (median value of the medium health range). Real-time available power of each unit was collected. The collected valid data were arithmetically averaged, and outliers exceeding 3σ were removed. The minimum available power average for medium dispatch was found to be 2.1kW. After calibration to 2kW to meet system dispatch redundancy requirements, and after 50 cluster dispatch tests verifying that this threshold could accurately distinguish medium power output units, a second power threshold of 2kW was pre-configured into the power judgment module of the optimized control unit. The second rated power ratio, based on the system's hierarchical power allocation strategy, is built using MATLAB / Simulink. This is based on the first rated power ratio of 80%, and takes into account the power requirements of medium-sized scheduling tasks. The simulation model simulates system scheduling response speed, unit operating load, and lifespan loss data under three candidate ratios: 30%, 40%, and 50%. Simultaneously, five groups of medium-state nodes are selected for actual operating condition verification. The scheduling adaptation effect of each ratio is statistically analyzed, and 40% is determined to be the optimal ratio (balancing scheduling requirements and unit safety). After multiple simulations verifying its feasibility, the second rated power ratio of 40% is pre-set in the power allocation algorithm module of the distributed cooperative controller.

[0046] Specifically, in this embodiment, the preset intermediate operating range is set to SOC 30%~60%, the preset medium health range is set to SOH 70%~90%, the second power threshold is set to 2kW, the first rated power ratio is set to 80%, and the second rated power ratio is set to 40%. The distributed collaborative controller receives precise status parameters uploaded by a distributed control node. The node's calculated SOC is 50%, its calculated SOH is 85%, its real-time available power is 3kW, and its rated power is 5kW. The distributed collaborative controller compares the node's SOC with the preset intermediate operating range. The system compares the SOC (State of Charge) with the preset medium health range of 70% to 90%, and the real-time available power with the first power threshold of 4kW and the second power threshold of 2kW. If the node's SOC is within the preset medium operating range, its SOH is within the preset medium health range, and its real-time available power is less than the first power threshold and greater than the second power threshold, it meets all the judgment conditions. Then, through distributed control commands, the system allocates a scheduling power to the node that is less than 80% of its rated power and greater than 40% of its rated power. The corresponding allocated scheduling power is in the range of 2kW to 4kW.

[0047] Understandably, the differentiated power allocation strategy has been further improved, and a moderate power allocation rule has been formulated for nodes in a moderate operating state. Through precise quantitative conditions, energy storage units with a moderate State of Charge (SOC), a moderate State of Health (SOH), and a moderate power output capability are identified, and dispatch power between the first and second proportions of the rated power is allocated to them. This achieves precise power matching for energy storage units in a moderate state, which not only fully explores and utilizes the actual operating capabilities of such nodes, but also avoids overloading them. This improves the hierarchical power allocation system of the energy storage cluster and significantly enhances the rationality and comprehensiveness of the cluster's power allocation.

[0048] In some embodiments of this application, when formulating differentiated power allocation strategies for each control node in the cluster based on the precise status parameters uploaded by each control node, and matching the corresponding scheduling power according to the actual operating capacity of each control node, the method further includes: When the SOC is lower than the preset lower threshold or higher than the preset upper threshold, and the SOH is less than the preset health threshold and the real-time available power is less than or equal to the second power threshold, then the current node is allocated scheduling power less than or equal to the second proportion of the rated power through distributed control instructions.

[0049] Specifically, in this embodiment, the preset lower threshold is the minimum SOC threshold preset by the system. It is a criterion for determining that the remaining power of the energy storage unit is in an operating state that is too low and is close to the over-discharge protection threshold. It is used to identify energy storage nodes with insufficient power and potential operational risks, avoiding accelerated aging of energy storage units or safety hazards caused by excessively low power. The preset upper threshold is the maximum SOC threshold preset by the system. It is a criterion for determining that the remaining power of the energy storage unit is in an operating state that is too high and is close to the overcharge protection threshold. It is used to identify energy storage nodes with oversaturated power and that do not conform to normal scheduling conditions, avoiding impact on charging and discharging flexibility and single-unit operation due to excessively high power. The lifespan of the battery is considered. In this embodiment, the preset lower threshold is based on the safety operation specification for electrochemical energy storage unit cells (GB / T36547-2018 "Performance Test Method for Electrochemical Energy Storage System Energy Storage Units" Clause 5.4 Low Charge Protection) and the industry standard for low charge over-discharge protection (over-discharge protection SOC not less than 15% of rated capacity). This is combined with the charge characteristics of aged energy storage units (the low charge discharge boundary of aged units is 5% lower than that of healthy units) and the safe operation boundary. Eight groups of aged energy storage units (internal resistance > 21.8mΩ, cycle life > 2000 times) are selected, and the CT-4008 battery testing system is used to conduct low charge protection tests. The discharge test was conducted with a discharge rate of 0.2C and an ambient temperature of 25±2℃. Real-time data were collected for each unit, including terminal voltage, remaining charge, discharge current, and operating temperature. When the unit terminal voltage dropped to 3.0V (the lower limit of safe discharge voltage for the energy storage cell), the discharge current fluctuation was ≤0.1A, and the operating temperature showed no abnormal increase, the SOC value at this point was recorded as the minimum safe SOC for that unit. Each unit was tested six times, resulting in 48 data sets. Outliers exceeding the 3σ range (such as <25% or >35% of outliers) were removed using the Grubbs criterion. The remaining valid data were then arithmetically averaged to obtain a mean of 30%. Combined with the system's low-power emergency protection strategy (reserving 5% emergency power), ±1% debugging and calibration were performed. Three candidate thresholds of 29%, 30%, and 31% were tested respectively. After 50 verifications under actual working conditions, it was ensured that the thresholds could effectively avoid the risk of over-discharge, prevent unit damage, and have a false judgment rate of <1%. 30% was then determined as the preset lower limit threshold. This threshold was pre-written into the EEPROM storage module of the distributed collaborative controller and the Flash threshold storage unit of each local distributed control node. A threshold comparison trigger mechanism was set to ensure accurate determination of the low-power state. The preset upper limit threshold is based on the overcharge protection requirements of the electrochemical energy storage unit cells (overcharge warning voltage ≤3).Considering the characteristics of 65V and long-term operating life (capacity degradation rate increases significantly when SOC > 90%), and the power limit control requirements of energy storage system cluster scheduling (to avoid the inability to quickly respond to scheduling commands after the unit is fully charged), 6 groups of power-type energy storage units (5kWh / 10kW) and 6 groups of energy-type energy storage units (10kWh / 5kW) were selected. Overcharge tests were conducted using a CT-4008 battery testing system. The charging rate was set to 0.3C and the ambient temperature to 25±2℃. Real-time data on the terminal voltage, SOC, charging current, and cell temperature of each unit were collected. When the unit terminal voltage rose to 3.65V (overcharge warning voltage), the charging current dropped to 0.05C, and the cell temperature did not rise significantly, the SOC value at this time was recorded as... The highest safe State of Charge (SOC) of this unit was tested repeatedly five times per unit, collecting a total of 60 data sets. Origin software was used for data fitting and analysis. Abnormal data indicating overcharge risk at SOC > 95% were removed, and the mean of the valid data was calculated to be 92%. Considering the system cluster scheduling flexibility requirements (reserve 8% power for rapid response), the calibration was adjusted to 90%. Overcharge risk was simulated using PSCAD / EMTDC simulation software under extreme ambient temperatures of 15℃ and 35℃. After verifying that this threshold could 100% avoid overcharge risks without affecting scheduling flexibility, 90% was set as the preset upper limit threshold and pre-configured in the parameter determination module of the optimization and control unit. This was linked with the real-time SOC acquisition module to ensure accurate control of the power limit.

[0050] Specifically, in this embodiment, the preset lower threshold is set to SOC 30%, the preset upper threshold is set to SOC 90%, the preset health threshold is set to SOH 90%, the second power threshold is set to 2kW, and the second ratio of rated power is set to 40%. The rated power of the control node is 5kW. The distributed collaborative controller receives precise status parameters uploaded by a distributed control node. The calculated SOC value of this node is 25%, the calculated SOH value is 85%, and the calculated real-time available power value is 1.5kW. The distributed collaborative controller compares the node's SOC with the preset lower threshold of 30% and the preset upper threshold of 90%, and determines that the SOC is lower than the preset lower threshold. It compares the SOH with the preset health threshold of 90%, and determines that the SOH is less than the preset health threshold. It compares the real-time available power with the second power threshold of 2kW, and determines that the real-time available power is less than or equal to the second power threshold. All the determination conditions are met, and then the distributed control command allocates a scheduling power of less than or equal to 40% of its rated power to the node, corresponding to an allocated scheduling power of less than or equal to 2kW.

[0051] Understandably, a safety-oriented low-proportion power allocation strategy was developed for aging nodes with abnormal operating conditions. This strategy accurately identifies energy storage units with abnormal SOC, SOH below the health threshold, or insufficient real-time available power through quantitative conditions, and allocates scheduling power less than or equal to the second proportion of the rated power to them. This strictly limits the scheduling load of aging and abnormal units from the power regulation perspective, completely avoids unreasonable overload scheduling of aging units, effectively slows down the performance degradation rate of aging units, extends their service life, and also prevents operational failures caused by improper scheduling of aging units. From the power regulation perspective, this ensures the operational safety and stability of the entire distributed energy storage system.

[0052] Reference Figure 2 As shown in some embodiments of this application, a method for evaluating state data and optimizing cluster control of a distributed energy storage system includes: Each energy storage unit is treated as an independent local distributed control node. The operating status data of each control node is collected in real time. The operating status data includes response time, rated capacity, internal resistance, number of cycles, remaining power, terminal voltage, charging and discharging current and operating temperature. The node type of each control node is determined based on the operating status data. Based on the node type, different status data evaluation methods are assigned to each control node, and the precise status parameters of each control node are collected and calculated in real time based on the status data evaluation methods. The precise status parameters include SOC, SOH and real-time available power. The real-time available power includes maximum available power, continuous available output power and limited available power. Each control node uploads the locally calculated precise status parameters to the distributed collaborative controller. The distributed collaborative controller formulates differentiated power allocation strategies for each control node in the cluster based on the accurate status parameters uploaded by each control node, and matches the corresponding scheduling power according to the actual operating capacity of each control node.

[0053] It is understandable that the above-mentioned method and system for evaluating state data and optimizing cluster control of distributed energy storage systems have the same beneficial effects, and will not be elaborated further here.

[0054] It should be noted that: Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. In some instances, well-known structures and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0055] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of this application and form different embodiments.

[0056] The above description is merely a preferred embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A distributed energy storage system state data assessment and cluster optimization control system, characterized in that, include: The data acquisition unit is used to collect the operating status data of each control node in real time, with each energy storage unit as an independent local distributed control node. The operating status data includes response time, rated capacity, internal resistance, number of cycles, remaining power, terminal voltage, charging and discharging current and operating temperature, and determines the node type of each control node based on the operating status data. The data evaluation unit is used to assign different status data evaluation methods to each control node based on the node type, and to collect and calculate the precise status parameters of each control node in real time based on the status data evaluation methods. The precise status parameters include SOC, SOH and real-time available power. The real-time available power includes maximum available power, continuous available output power and limited available power. Each control node uploads the locally calculated precise status parameters to the distributed collaborative controller. An optimized control unit is used by the distributed collaborative controller to formulate differentiated power allocation strategies for each control node in the cluster based on the accurate status parameters uploaded by each control node, and to match the corresponding scheduling power according to the actual operating capacity of each control node.

2. The distributed energy storage system state data assessment and cluster optimization control system according to claim 1, characterized in that, When determining the node type of each control node based on the aforementioned operational status data, the following steps are included: When the response time is less than or equal to a preset millisecond-level response threshold, the rated capacity is less than or equal to a preset small capacity threshold, the internal resistance is less than or equal to a preset healthy internal resistance threshold, and the number of cycles is less than or equal to a preset healthy cycle number threshold, then the node type of the current control node is a distributed power-type healthy small capacity control node. When the rated capacity is greater than the preset small capacity threshold, the response time is greater than the preset millisecond response threshold, and the number of cycles is within the preset medium health cycle number range, then the node type of the current control node is a distributed energy type medium capacity medium health control node.

3. The distributed energy storage system state data assessment and cluster optimization control system according to claim 2, characterized in that, When determining the node type of each control node based on the aforementioned operational status data, the method further includes: When the internal resistance is greater than the preset healthy internal resistance threshold, the number of cycles is greater than the preset healthy cycle number threshold, and the remaining power is less than the preset remaining power threshold, then the node type of the current control node is a distributed decay type old control node.

4. The distributed energy storage system state data assessment and cluster optimization control system according to claim 3, characterized in that, When assigning different state data evaluation methods to each control node based on the node type, and when collecting and calculating the precise state parameters of each control node in real time based on the state data evaluation methods, the process includes: When the node type is a distributed power-type healthy small-capacity control node, a local state data evaluation method emphasizing fast response is adopted, and the current SOC, SOH and maximum available power of the current control node are obtained through local distributed calculation using the terminal voltage, the charging and discharging current and the operating temperature.

5. The distributed energy storage system state data assessment and cluster optimization control system according to claim 4, characterized in that, When assigning different state data evaluation methods to each control node based on the node type, and when collecting and calculating the precise state parameters of each control node in real time based on the state data evaluation methods, the method further includes: When the node type is a distributed energy type medium capacity medium health control node, a local energy-based state data evaluation method is adopted, and the SOC, SOH and continuous available output power of the current control node are obtained by local distributed calculation using the cumulative charge and discharge amount, the remaining power, the actual available capacity and the operating temperature.

6. The distributed energy storage system state data assessment and cluster optimization control system according to claim 5, characterized in that, When assigning different state data evaluation methods to each control node based on the node type, and when collecting and calculating the precise state parameters of each control node in real time based on the state data evaluation methods, the method further includes: When the node type is a distributed decay type old control node, a local state data evaluation method focusing on safety limitations is adopted, and the SOC, SOH and the limited available power of the current control node are obtained by local distributed calculation using internal resistance change, voltage deviation and the remaining power.

7. The distributed energy storage system state data assessment and cluster optimization control system according to claim 6, characterized in that, Based on the precise status parameters uploaded by each control node, a differentiated power allocation strategy is formulated for each control node in the cluster. When matching the corresponding scheduling power according to the actual operating capacity of each control node, the following is included: When the SOC is in a preset normal operating range, the SOH is greater than or equal to a preset health threshold, and the real-time available power is greater than a first power threshold, then a scheduling power greater than or equal to a first proportion of the rated power is allocated to the current node through distributed control instructions.

8. The distributed energy storage system state data assessment and cluster optimization control system according to claim 7, characterized in that, Based on the precise status parameters uploaded by each control node, a differentiated power allocation strategy is formulated for each control node in the cluster. When matching the corresponding scheduling power according to the actual operating capacity of each control node, the strategy also includes: When the SOC is in a preset intermediate operating range and the SOH is in a preset medium health range, and the real-time available power is less than a first power threshold and greater than a second power threshold, then a scheduling power less than a first proportion of the rated power and greater than a second proportion of the rated power is allocated to the current node through distributed control instructions.

9. The distributed energy storage system state data assessment and cluster optimization control system according to claim 8, characterized in that, Based on the precise status parameters uploaded by each control node, a differentiated power allocation strategy is formulated for each control node in the cluster. When matching the corresponding scheduling power according to the actual operating capacity of each control node, the strategy also includes: When the SOC is lower than the preset lower threshold or higher than the preset upper threshold, and the SOH is less than the preset health threshold and the real-time available power is less than or equal to the second power threshold, then the current node is allocated scheduling power less than or equal to the second proportion of the rated power through distributed control instructions.

10. A method for evaluating state data and optimizing cluster control of a distributed energy storage system, characterized in that, The distributed energy storage system state data assessment and cluster optimization control system according to any one of claims 1-9 includes: Each energy storage unit is treated as an independent local distributed control node. The operating status data of each control node is collected in real time. The operating status data includes response time, rated capacity, internal resistance, number of cycles, remaining power, terminal voltage, charging and discharging current and operating temperature. The node type of each control node is determined based on the operating status data. Based on the node type, different status data evaluation methods are assigned to each control node, and the precise status parameters of each control node are collected and calculated in real time based on the status data evaluation methods. The precise status parameters include SOC, SOH and real-time available power. The real-time available power includes maximum available power, continuous available output power and limited available power. Each control node uploads the locally calculated precise status parameters to the distributed collaborative controller. The distributed collaborative controller formulates differentiated power allocation strategies for each control node in the cluster based on the accurate status parameters uploaded by each control node, and matches the corresponding scheduling power according to the actual operating capacity of each control node.

Images