Modular direct current microgrid intelligent energy storage method and related device

Abstract

The application is suitable for the technical field of power grid energy storage control, and provides a modular DC micro-grid intelligent energy storage method and related equipment. The method comprises the following steps: when a new module is connected to a DC micro-grid, module information is collected to obtain a frequency domain power characteristic spectrum; cross-module correlation analysis is performed according to the frequency domain power characteristic spectrum to construct a micro-grid power topology graph; each device is traversed based on the micro-grid power topology graph, and module capacity attenuation prediction is performed to output an SOH prediction result set of each module; a charging instruction is issued to each energy storage module according to the SOH prediction result set, and charging mean deviation calculation is performed to obtain a charging mean deviation value of each module. The application improves the energy utilization rate, operation stability and equipment life of the modular DC micro-grid, and reduces operation and maintenance costs and fault risks.

Description

Technical Field

[0001] This invention relates to the field of power grid energy storage control technology, and in particular to a modular DC microgrid intelligent energy storage method and related equipment. Background Technology

[0002] With the continuous development of new energy power generation technology and power electronics technology, modular DC microgrids, as an efficient, flexible, and scalable form of energy organization, have been widely used in industrial parks, data centers, smart buildings, and distributed energy access scenarios. By unifying and coordinating the access of photovoltaic power generation, energy storage systems, DC loads, and power conversion devices, they can effectively improve the absorption capacity of renewable energy and reduce energy conversion losses. In the multi-module collaborative operation mode, the energy storage system, as the core regulation unit in the DC microgrid, plays a crucial role in balancing power fluctuations, stabilizing DC bus voltage, and improving the reliability of system power supply.

[0003] In actual operation, modular DC microgrids contain a large number of energy storage units of various types. Different energy storage modules exhibit significant differences in capacity, health status, response speed, and charge / discharge capabilities, leading to a prominent issue of uneven power distribution within the system. Furthermore, the inherent randomness and volatility of renewable energy generation cause energy storage modules to frequently switch between dynamic charge and discharge states, which can accelerate battery aging and cause capacity degradation. In addition, under high load or rapid power change scenarios, the internal temperature rise of energy storage modules is significant. Without effective thermal management mechanisms, this can lead to localized overheating and even affect the stable operation of the system. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention proposes a modular DC microgrid intelligent energy storage method and related equipment, thereby resolving at least one of the aforementioned technical issues.

[0005] In a first aspect, embodiments of this application provide a modular DC microgrid smart energy storage method, comprising the following steps: Step S1: When the new module is connected to the DC microgrid, the module information is collected to obtain the frequency domain power characteristic spectrum; Step S2: Perform cross-module correlation analysis based on the frequency domain power characteristic spectrum to construct a microgrid power topology map; Step S3: Based on the microgrid power topology map, traverse each device one by one, perform module capacity attenuation prediction, and output the SOH prediction result set for each module; Step S4: Based on the SOH prediction result set, issue charging instructions to each energy storage module and calculate the average charging deviation to obtain the average charging deviation value of each module. Step S5: Calculate the upper limit of charging and discharging power based on the average charging deviation value to obtain the upper limit of power carrying capacity of each module; Step S6: Calculate the liquid cooling power requirement and perform cooling control processing based on the upper limit of the power capacity of each module.

[0006] Secondly, embodiments of this application provide a modular DC microgrid smart energy storage device, comprising: The acquisition unit is used to acquire module information and obtain the frequency domain power characteristic spectrum when a new module is connected to the DC microgrid. Topology unit, used to perform cross-module correlation analysis based on the frequency domain power characteristic spectrum, and construct microgrid power topology map; The attenuation prediction unit is used to traverse each device based on the power topology map of the microgrid, predict the capacity attenuation of the modules, and output the SOH prediction result set of each module. The deviation calculation unit is used to issue charging instructions to each energy storage module based on the SOH prediction result set, and to calculate the average charging deviation to obtain the average charging deviation value of each module. The upper limit calculation unit is used to calculate the upper limit of charging and discharging power based on the average charging deviation value, so as to obtain the upper limit of power carrying capacity of each module. The cooling control unit is used to calculate the liquid cooling power requirement and control the cooling process based on the power capacity limit of each module.

[0007] Thirdly, embodiments of this application provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the above-described modular DC microgrid smart energy storage method.

[0008] In one solution provided by the aforementioned methods, devices, and equipment, by uniformly identifying, registering, and collecting electrical parameters of newly connected modules, and extracting frequency domain power characteristic spectra using short-time Fourier transform, complex time-domain power fluctuations can be converted into quantifiable frequency domain characteristic information. This improves the ability to identify module operating status, power change patterns, and abnormal characteristics, providing a reliable data foundation for refined microgrid management. Cross-module correlation analysis using frequency domain power characteristic spectra and the construction of a microgrid power topology map accurately reflects the energy interaction relationships and power coupling degree between modules, enabling a visual representation of complex microgrid structures and improving power flow analysis efficiency and module collaborative management capabilities. By performing a traversal analysis of each energy storage module and combining SOC estimation and capacity decay prediction to obtain SOH results, the health status and aging degree of each energy storage module can be accurately grasped, allowing for timely detection of modules with degraded performance, improving energy storage resource utilization efficiency and equipment operational reliability.

[0009] By allocating charging tasks based on the State of Health (SOH) and performing mean deviation analysis on the charging process, differences in charging consistency among energy storage modules can be identified, potential performance anomalies and capacity imbalances can be discovered, and quantitative basis for balanced control and optimized energy storage scheduling can be provided. By combining the charging mean deviation value to calculate the actual power carrying capacity of each module, safe charge and discharge boundaries can be dynamically determined, preventing aging modules from operating under high load for extended periods, improving the operational safety of the energy storage system, and extending the overall lifespan of the energy storage modules. By predicting heat load demand based on the power carrying capacity limit and implementing tiered liquid cooling control, the risk of temperature rise can be suppressed in advance, achieving precise heat dissipation management, reducing the impact of high temperatures on battery performance and lifespan, and improving the operational stability and safety of the modular DC microgrid under high load conditions. Attached Figure Description

[0010] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a schematic diagram of the system structure in one embodiment of the present invention; Figure 2 This is a flowchart illustrating a modular DC microgrid intelligent energy storage method according to an embodiment of the present invention; Figure 3 yes Figure 2 A schematic diagram of the implementation process of step S1; Figure 4 yes Figure 2 A schematic diagram of the implementation process of step S2; Figure 5 This is a schematic diagram of a modular DC microgrid intelligent energy storage device according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] It should be understood that, when used in this specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or collections thereof. It should also be understood that, as used in this specification and the appended claims, the term "and / or" refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0014] Furthermore, in the description of this invention and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0015] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of the invention include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0016] It should be understood that the sequence number of each step in the following embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0017] To illustrate the technical solution of the present invention, specific embodiments are described below.

[0018] To address the problems mentioned above in the background art, this application proposes a modular DC microgrid intelligent energy storage method and related equipment. The method provided by this invention can be applied to applications such as... Figure 1 The system shown includes a client and a server.

[0019] In one embodiment, such as Figure 2 As shown, a modular DC microgrid smart energy storage method is provided, which is applied to... Figure 1 Taking the system in the example, the following steps are included: Step S1: When the new module is connected to the DC microgrid, the module information is collected to obtain the frequency domain power characteristic spectrum; Step S2: Perform cross-module correlation analysis based on the frequency domain power characteristic spectrum to construct a microgrid power topology map; Step S3: Based on the microgrid power topology map, traverse each device one by one, perform module capacity attenuation prediction, and output the SOH prediction result set for each module; Step S4: Based on the SOH prediction result set, issue charging instructions to each energy storage module and calculate the average charging deviation to obtain the average charging deviation value of each module. Step S5: Calculate the upper limit of charging and discharging power based on the average charging deviation value to obtain the upper limit of power carrying capacity of each module; Step S6: Calculate the liquid cooling power requirement and perform cooling control processing based on the upper limit of the power capacity of each module.

[0020] In this embodiment, after the energy storage module, photovoltaic module, bidirectional DC / DC converter module, or DC load module is connected to the DC microgrid, it completes identity registration by sending a device description frame via the CAN bus. The device description frame includes information such as device type, rated power, protocol version, and current SOC, where the rated power range is maintained at 100 W to 100 kW, and the SOC value range is maintained at 0% to 100%. After registration is completed, electrical parameters of each module are collected according to the module registration information table. The collected content includes bus voltage, battery terminal voltage, output current, output power, and operating status information. The voltage sampling range is maintained at 0 V to 1000 V, the current sampling range is maintained at 0 A to 500 A, and the sampling frequency is maintained at 1 kHz, which is increased to 5 kHz in scenarios with large dynamic power fluctuations. Subsequently, abnormal sampling point removal and missing value repair processing are performed on the collected original voltage and current sampling sequences to form a continuous electrical sampling sequence. Then, short-time Fourier transform was used to perform time-frequency analysis on the continuous electrical sampling sequence. The window length was set to 256 to 1024 points, preferably 512 points, and the overlap rate of adjacent windows was set to 25% to 75%, preferably 50%. By analyzing the energy distribution in different frequency ranges, characteristic parameters such as the dominant frequency power component, harmonic energy distribution, and frequency band energy proportion were extracted, ultimately forming a frequency domain power characteristic spectrum, providing basic data for module correlation analysis and topology construction.

[0021] Using the frequency domain power characteristic spectrum corresponding to each module as the analysis object, parameters such as frequency energy distribution, main frequency fluctuation amplitude, and harmonic response characteristics are extracted to construct the module frequency domain feature vector. Subsequently, correlation analysis is performed on the frequency domain characteristics between multiple modules to calculate the power coupling strength coefficient between modules. The power coupling strength coefficient ranges from 0 to 1, with values ​​closer to 1 indicating a higher degree of power correlation between modules. A module correlation matrix is ​​established based on the power coupling strength coefficient, and each module is considered a topology node, mapping the power coupling relationship between modules to node connection edges. A strong correlation is determined when the power coupling strength coefficient is higher than a preset threshold, which is set to 0.6–0.8, preferably 0.7. Then, topological clustering is performed based on the electrical connection relationship, power flow relationship, and correlation strength between modules to form multiple power collaboration domains. Finally, a microgrid power topology map is established by combining equipment type, bus connection relationship, and power transmission path, enabling the topology map to intuitively reflect the energy flow relationship and collaborative operation status between energy storage modules, power generation modules, and load modules.

[0022] Based on the microgrid power topology diagram, all device nodes are traversed one by one to select energy storage module nodes, and corresponding operating parameters are extracted, including battery terminal voltage, bus voltage, charging and discharging current, output power, battery temperature, cumulative charging and discharging capacity, and cycle count. The voltage sampling range is maintained at 0 V to 1000 V, and the current sampling range is maintained at 0 A to 500 A. Subsequently, a State of Charge (SOC) estimation model is established using the charging and discharging current, battery terminal voltage, and rated capacity, and the SOC is dynamically corrected using an extended Kalman filter to obtain estimated SOC values ​​for each module, with SOC values ​​ranging from 0% to 100%. Then, the estimated SOC values, along with parameters such as battery temperature, cumulative cycle count, and cumulative charging and discharging capacity, are input into a pre-set deep learning network to analyze the capacity decay trend of the energy storage modules. By learning the mapping relationship between historical operating states and capacity decay, the current State of Health (SOH) of each module is predicted, with SOH values ​​ranging from 0% to 100%. Finally, a set of SOH prediction results is formed, and a mapping relationship is established between this prediction and the unique identifier of the energy storage module, providing a basis for energy storage scheduling and lifespan management.

[0023] Based on the SOH (State of Health) prediction results for each energy storage module, the module health status is graded and assessed. A module with an SOH above 90% is considered to be in high health; a module with an SOH between 70% and 90% is considered to be in normal health; and a module with an SOH below 70% is considered to be in a capacity-degrading state. Charging tasks are assigned according to the health level, SOC (State of Charge) status, and rated capacity, and charging control commands are sent to each energy storage module. During charging, a gradual current ramp-up method is used for charging control, while charging current information is collected in real-time at a sampling frequency of 1 kHz, increasing to 5 kHz in scenarios with large power variations. The collected information includes the initial current value, the current ramp-up process, the current stability value, and the ramp-up duration, forming a current ramp-up process record. Subsequently, the average current ramp-up record of each energy storage module is calculated to obtain the corresponding average charging current. The deviation between this average current and the overall module average charging current is then calculated to form the average charging deviation value. Mean deviation analysis can identify consistency differences and capacity degradation levels among energy storage modules, providing data for balanced control.

[0024] Active balancing analysis is performed based on the average charging deviation of each energy storage module, and energy balancing adjustment is combined with bidirectional DC / DC converters. When the average deviation is large, the consistency difference is reduced by adjusting the energy distribution ratio between modules. After balancing adjustment, the actual usable capacity of each energy storage module is calculated by combining the estimated SOC, predicted SOH, rated capacity, and actual operating status. The SOC and SOH values ​​are maintained within the range of 0% to 100%, and the rated power range is maintained within the range of 100W to 100kW. Subsequently, a power capacity assessment model is established based on the actual usable capacity, current state of charge, and health status to calculate the maximum charging power and maximum discharging power of each module. When the SOC is close to 100%, the upper limit of charging power is reduced; when the SOC is close to 0%, the upper limit of discharging power is reduced; for modules with low SOH, the power carrying capacity is reduced simultaneously. Finally, the upper limits of charging power and discharging power for each energy storage module are generated, forming the module power carrying capacity upper limit results, providing a basis for power prediction and thermal management control.

[0025] Based on the power capacity limit of each energy storage module, the charging and discharging power demand during the future operating cycle is predicted, and the corresponding heat load level is calculated in conjunction with the power change trend. Subsequently, a heat load analysis model is established based on the predicted power value, battery temperature, current changes, and historical temperature rise patterns to generate a set of module temperature rise prediction curves and calculate the corresponding liquid cooling power demand. An intelligent cooling control strategy is executed based on the liquid cooling power demand. When the prediction indicates a risk of temperature rise in the module within the next 5 to 10 minutes, liquid cooling pre-cooling is initiated in advance. A graded thermal management mechanism is adopted during the control process: when the module temperature is below 30℃, only air cooling mode is activated; when the temperature is between 30℃ and 35℃, low-flow liquid cooling is used in conjunction with air cooling; when the temperature exceeds 35℃, the liquid cooling system enters full-power operation and the fan speed is increased; when the temperature exceeds 40℃, both liquid cooling and air cooling operate at full power simultaneously, triggering the EMS power reduction protection mechanism. During the control process, the liquid cooling pump speed, coolant flow rate, fan speed and temperature changes are continuously recorded to form a thermal management response execution record. The record is then evaluated and archived based on the temperature convergence to generate an energy storage thermal state management execution record, thereby improving the safety, stability and lifespan of the modular DC microgrid energy storage system.

[0026] In one embodiment, such as Figure 3 As shown, step S1 specifically includes the following steps: When a new module is connected to a DC microgrid, it broadcasts a device description frame via the CAN bus. Handshake confirmation is performed based on the device description frame, and the data is incorporated into the DC microgrid topology management to generate a module registration information table. Electrical parameters are scanned based on the module registration information table to obtain the original voltage and current sampling sequence; The original voltage and current sampling sequence is subjected to abnormal sampling point removal and missing value interpolation repair to obtain a continuous electrical sampling sequence; A short-time Fourier transform is performed on the continuous electrical sampling sequence to obtain the frequency domain power characteristic spectrum.

[0027] In this embodiment, when an energy storage module, photovoltaic module, load module, or converter module is connected to the DC microgrid, the module control unit completes a power-on self-test and starts the CAN communication interface. After initialization, it periodically sends device description frames via the CAN bus to declare its identity and operational capabilities to other nodes in the microgrid. The device description frame is encapsulated using the standard CAN data format. The device type field identifies the current module category, the rated power field characterizes the module's designed output capability (ranging from 100 W to 100 kW), the protocol version field identifies communication compatibility, and the current SOC field characterizes the remaining power status of the energy storage module (ranging from 0% to 100%). The broadcast period for the device description frame is set to 100 ms to 500 ms, preferably 200 ms, to ensure that newly connected modules can be quickly discovered by the network. During the broadcast, auxiliary information such as the module's unique identification code, device operating status, and fault indicators are simultaneously included, enabling nodes in the network to accurately identify the module's identity and current operational capabilities, providing basic data support for module access management and resource scheduling. When a microgrid control node receives a device description frame, it verifies the validity of the device type, protocol version, and device identification information. If the protocol version matches and the device status is normal, a handshake confirmation message is sent to the newly connected module to establish a communication connection. During the handshake process, node address allocation, communication permission confirmation, and functional parameter synchronization are performed to ensure the module can participate in microgrid operation according to unified communication rules. After the handshake confirmation, a node file is created based on the module's unique identifier, recording information such as device type, rated power, SOC status, access time, and communication status. The newly connected module is then incorporated into the current DC microgrid topology. A topology mapping relationship is constructed based on the electrical connections between nodes, forming a module registration information table. This table records not only the basic attributes of each node but also the connection relationships between nodes and their respective functional areas, thus forming a complete microgrid device management list. This enables unified management and dynamic maintenance of newly added modules, improving the scalability and access compatibility of the modular DC microgrid.

[0028] Based on the node information in the module registration information table, periodic electrical parameters are collected from registered modules. The collected parameters include module port voltage, output current, power value, bus voltage, and energy storage unit charging / discharging status. The voltage sampling range can be set to 0 V–1000 V, and the current sampling range can be set to 0 A–500 A. The sampling frequency is set according to the microgrid's operating status, set to 1 kHz under normal operating conditions, and can be increased to 5 kHz in scenarios with large dynamic power fluctuations. During the sampling process, a synchronous clock mechanism is used to unify the sampling time of each module, reducing the impact of multi-node sampling time deviations on the analysis results. Each sampling is accompanied by a corresponding timestamp and stored in chronological order. The continuously collected voltage and current data form the original voltage and current sampling sequence, which accurately reflects the electrical changes during module operation, providing a basic data source for subsequent operating status analysis, power fluctuation identification, and fault detection.

[0029] Data quality processing is performed on the acquired raw voltage and current sampling sequences. A sliding window statistical method is used to detect anomalies in the sampling data, with the statistical window length set to 50 to 200 sampling points, preferably 100. The mean and standard deviation are calculated within each window. A sampling point deviating from the window mean by more than three times the standard deviation is identified as an anomaly. Causes of anomalies include communication interference, sensor transient distortion, and sampling jitter. Identified anomalies are removed. Subsequently, data gaps caused by communication interruptions or sampling anomalies are repaired. When the gap length is less than 10 consecutive sampling points, linear interpolation is used for compensation; when the gap length reaches more than 10 consecutive sampling points, trend interpolation is used to repair the gap by considering the trend of changes in the preceding and following valid data. After anomaly removal and gap repair, a continuous electrical sampling sequence is obtained, ensuring the voltage and current curves maintain temporal continuity and completeness of changes, thus improving the accuracy and stability of electrical feature analysis.

[0030] Time-frequency analysis is performed on continuous electrical sampling sequences to identify power variation characteristics corresponding to different frequency components during microgrid operation. The continuous electrical sampling sequences are segmented according to preset time windows, with window lengths ranging from 256 to 1024 points, preferably 512 points; the overlap rate between adjacent windows is set to 25%–75%, preferably 50%. Subsequently, short-time Fourier transform is used to perform frequency domain decomposition on the voltage and current signals within each time window to obtain the amplitude and energy distribution corresponding to different frequency components. Based on the voltage and current spectra in the frequency domain, the frequency domain power distribution characteristics are further calculated to obtain the frequency domain power characteristic spectrum. The frequency domain power characteristic spectrum can reflect the frequency characteristics generated by load fluctuations, energy storage charging and discharging switching, and changes in the operating state of power electronic equipment during module operation. It can also identify phenomena such as harmonic enhancement, power oscillation, and abnormal frequency components, providing a reliable basis for module operating status assessment, energy dispatch optimization, and fault early warning analysis.

[0031] In one embodiment, such as Figure 4 As shown, step S2 specifically includes the following steps: Identify information from multiple modules in a DC microgrid; Based on the frequency domain power characteristic spectrum, cross-module correlation analysis is performed on the information of multiple modules to obtain the power coupling strength coefficient between different modules; Based on the power coupling strength coefficient, topological clustering is performed to obtain the power cooperative domain partitioning result; Communication configuration is performed based on the power coordination domain partitioning results, and the built-in protocol of the adaptation layer is parsed to construct the microgrid power topology diagram.

[0032] In this embodiment, various modules in the DC microgrid are uniformly identified and classified based on a previously generated module registration information table. The identified objects include energy storage modules, photovoltaic power generation modules, bidirectional converter modules, DC load modules, and energy management and control modules. Basic information such as the device's unique identifier, device type, rated power, current SOC, communication status, and access time are read from the module registration information table, and combined with real-time collected voltage, current, and power parameters to establish a module operation profile. The rated power range of the modules is maintained between 100 W and 100 kW, and the SOC range is 0% to 100%. Simultaneously, based on the module's connection relationship in the topology, its bus location and upstream and downstream associated nodes are identified, forming a module association set. For modules added or removed during operation, dynamic updates are performed via CAN bus status broadcast information to ensure consistency between the identification results and the actual operating status. After identification, the basic attributes, operating status parameters, and electrical connection relationships of each module are uniformly organized to form a module information set, providing basic data support for cross-module power relationship analysis and topology reconfiguration.

[0033] Cross-module correlation analysis is performed using the frequency domain power characteristic spectra of each module to assess the degree of power interaction and energy synergy between different modules. Characteristic parameters such as the dominant frequency power component, harmonic energy distribution, and frequency band energy proportion are extracted from the frequency domain power characteristic spectra, and corresponding frequency domain power characteristic vectors are established. Subsequently, multiple module frequency domain characteristic vectors acquired synchronously in time are used as the analysis object, and correlation analysis is used to calculate the frequency response consistency and power fluctuation synchronization between modules. When two modules exhibit similar power change characteristics in the same frequency range, a strong power coupling relationship is determined between them. Based on the calculation results, a power coupling strength coefficient between modules is generated. The power coupling strength coefficient ranges from 0 to 1, where a value close to 0 indicates a weak correlation between modules, and a value close to 1 indicates a strong power synergy between modules. This analysis process can identify the power interaction patterns between energy storage modules and load modules, photovoltaic modules and energy storage modules, and different converter modules, providing a quantitative basis for microgrid energy flow analysis and coordinated control.

[0034] A power correlation matrix is ​​constructed based on the power coupling strength coefficient between modules, and this matrix serves as the basis for topological clustering. Each module is treated as an independent node, and the power coupling strength coefficient is used as the connection weight between nodes, forming a module power correlation network. Clustering analysis is then performed based on the correlation weights. When the power coupling strength coefficient between modules exceeds a preset threshold, they are assigned to the same power coordination domain. This threshold can be set to 0.6–0.8, preferably 0.7. Modules with weaker coupling relationships are assigned to different coordination domains. The clustering process comprehensively considers factors such as module type, rated power, and actual power flow direction, ensuring that the partitioning results reflect both frequency domain power correlation and actual electrical connection characteristics. After clustering, multiple relatively independent power coordination domains are formed, with modules within each domain exhibiting strong energy exchange relationships and operational coordination. Power coordination domain partitioning reduces the overall analysis complexity of large-scale modular microgrids and improves the targeting and execution efficiency of energy management and power scheduling.

[0035] Based on the obtained power coordination domain division results, the internal communication resources of the microgrid are optimized. Priority communication relationships are established for modules within the same power coordination domain to increase the data exchange frequency between highly correlated modules; for communication between different coordination domains, a periodic interactive method is used for state synchronization. The communication period can be set to 100 ms to 500 ms, preferably 200 ms, consistent with the communication period during the module access phase. Subsequently, the built-in protocol parsing mechanism of the adaptation layer is activated to uniformly parse and convert the CAN bus messages, device status messages, and power scheduling commands, enabling different types of modules to exchange information according to a unified data structure. After completing the communication relationship mapping, a node connection model is established based on the electrical connection relationships, power flow relationships, and power coupling strength coefficients between modules, mapping each module as a topology node and the power correlation relationships between nodes as topology connection edges. The weight of the connection edge is represented by the power coupling strength coefficient, whose value range is maintained between 0 and 1. By constructing a microgrid power topology map, the energy flow paths, power coordination relationships, and key power node distribution between modules can be intuitively displayed, providing visualized data support for energy storage scheduling optimization, power balance control, and fault location analysis.

[0036] In one embodiment, step S3 specifically includes the following steps: Based on the power topology diagram of the microgrid, each device is traversed one by one to extract the operating parameters of each energy storage module; The SOC estimation and extended Kalman filtering are performed on the operating parameters to obtain the SOC estimation value of each module; The estimated SOC value is input into a preset deep learning network to predict module capacity decay and output the SOH prediction result set for each module.

[0037] In this embodiment, based on the constructed microgrid power topology map, all node devices in the topology map are identified one by one, and energy storage module nodes are selected according to device type. For the identified energy storage modules, the basic device information in the module registration information table is read, including parameters such as the device's unique identifier, device type, rated power, current SOC, communication status, and access time. The rated power range is maintained at 100 W to 100 kW, and the SOC value range is maintained at 0% to 100%. Subsequently, combined with real-time collected electrical operation data, the operating parameters of the energy storage module are extracted. The operating parameters include battery terminal voltage, bus voltage, charging and discharging current, output power, battery temperature, cumulative charging capacity, cumulative discharging capacity, and cycle count. The voltage sampling range is maintained at 0 V to 1000 V, the current sampling range is maintained at 0 A to 500 A, and the sampling frequency is maintained at 1 kHz, which is increased to 5 kHz in scenarios with large dynamic power fluctuations. To ensure the time consistency between parameter data, various operating parameters are synchronously associated according to a unified timestamp, and an operating parameter sequence is established according to the time order. For abnormal data and short-term missing data that occur during the sampling process, the aforementioned abnormal sampling point removal and interpolation repair mechanism is used to ensure the continuity and integrity of the operating parameter sequence. After traversal and extraction, a dataset of operating parameters for each energy storage module is formed, providing basic input data for SOC estimation and health status analysis.

[0038] The State of Charge (SOC) of the battery is dynamically estimated using the operating parameters of the energy storage module. The charge change per unit time is calculated based on the collected charge and discharge current data, and an SOC change model is established in conjunction with the rated capacity of the energy storage module to obtain an initial SOC estimate. Simultaneously, data on battery terminal voltage, battery temperature, and power changes at corresponding times are read to constrain and correct the SOC estimate, reducing estimation deviations caused by accumulated current integration errors. Since the energy storage module is affected by sensor noise, battery polarization effects, and load fluctuations during actual operation, an extended Kalman filter (EPF) method is introduced for dynamic optimization of the SOC. The EPF predicts the current SOC using the battery state equation, then calculates the prediction error based on the real-time measured battery terminal voltage, and corrects and updates the SOC result according to the error magnitude. A prediction and correction process is performed once per sampling period, ensuring that the SOC continuously tracks the actual state of charge changes of the battery. After processing with the EPF, the influence of measurement noise and parameter drift on the estimation results can be effectively suppressed, improving the accuracy and stability of the SOC estimation. The estimated SOC value for each energy storage module is obtained, and the estimated SOC value is associated with the module's unique identifier and stored to form an energy storage status assessment dataset.

[0039] The estimated State of Charge (SOC) of each energy storage module is used as the core input feature, combined with operational parameter data, and input into a pre-defined deep learning network for capacity degradation prediction and analysis. Input features include SOC estimates, battery terminal voltage, charge / discharge current, battery temperature, cumulative charge / discharge capacity, cycle count, and historical operating status. These multi-dimensional features collectively reflect the current operating status and long-term aging characteristics of the energy storage module. The deep learning network utilizes the mapping relationship between historical operating samples and capacity degradation samples to learn and predict the capacity change trend of the energy storage module, identifying degradation patterns under different operating conditions. During the prediction process, the network analyzes the remaining usable capacity change trend of the energy storage module based on the current SOC changes and long-term operating characteristics, and calculates the corresponding State of Health (SOH) index. SOH characterizes the health level of the energy storage module, with a value range of 0% to 100%. When SOH approaches 100%, the energy storage module is in good condition; when SOH gradually decreases, it indicates that the module capacity is degrading and the degree of aging is increasing. After the prediction is completed, the SOH prediction results for each energy storage module are output, and a corresponding relationship is established based on the device's unique identifier, forming an SOH prediction result set. By conducting a unified assessment and comparative analysis of the health status of multiple energy storage modules, energy storage units with abnormal capacity decay or rapid lifespan decline can be identified in a timely manner. This provides data support for energy storage module maintenance and replacement decisions, as well as energy dispatch optimization, thereby improving the reliability and stability of the modular DC microgrid energy storage system.

[0040] In one embodiment, step S4 specifically involves the following steps: Based on the SOH prediction result set, the energy storage capacity of each module is allocated to generate an energy storage allocation scheme; Based on the energy storage distribution scheme, charging commands are issued to each energy storage module, and charging current information is collected to generate a record of the current ramp-up process. The average charging value of each module is calculated based on the current ramp-up process record, and the mean deviation is calculated to obtain the average charging deviation value of each module.

[0041] In this embodiment, energy storage resources are optimized and configured based on the SOH prediction result set corresponding to each energy storage module. Parameters such as the predicted SOH value, estimated SOC value, rated power, and rated capacity of each energy storage module are read, with SOH values ​​ranging from 0% to 100%, SOC values ​​ranging from 0% to 100%, and rated power ranging from 100 W to 100 kW. Subsequently, a health level assessment is performed on each energy storage module. Modules with an SOH higher than 90% are classified as high-health modules; those with an SOH between 70% and 90% are classified as normal-health modules; and those with an SOH lower than 70% are classified as capacity-degrading modules. During capacity allocation, the energy-bearing ratio of high-health modules is increased, while the charging and discharging load ratio of capacity-degrading modules is reduced, thereby mitigating further wear and tear on aging modules. Simultaneously, capacity coordination is performed based on the current SOC status; when the SOC is high, the charging allocation ratio is appropriately reduced, and when the SOC is low, the allocable charging capacity is increased. Based on the health status, remaining capacity, and power capability of each module, the corresponding energy storage allocation weight is calculated, and the total energy storage capacity task is allocated according to the weight ratio to form an energy storage allocation scheme. The energy storage allocation scheme includes the target charging capacity, target power range, and expected participation ratio of each module, so that energy storage resources can be rationally utilized according to their health status and operational capabilities, thereby improving the overall operating efficiency and service life of the energy storage system.

[0042] According to the generated energy storage allocation scheme, corresponding charging control commands are sent to each energy storage module. These commands include parameters such as the target charging current, target charging power, and charging execution duration. Upon receiving the control command, each energy storage module initiates the charging process according to a predetermined charging strategy. During the charging start-up phase, to avoid instantaneous high current impacting the battery and bus system, a gradual current ramp-up method is adopted for charging control; that is, the charging current gradually increases to the target value according to a set slope. During the charging process, charging current data for each module is collected in real time, with the sampling frequency maintained at 1 kHz, increased to 5 kHz in scenarios with large power variations, and corresponding timestamp information is recorded. The collected data includes parameters such as the initial current value, the current growth process, the target current value when the current reaches a stable state, and the duration of the entire ramp-up phase. For each energy storage module, all current change data from charging start-up to reaching a stable charging state are recorded in chronological order, forming a corresponding current ramp-up process record. By recording the dynamic response of different energy storage modules during the charging start-up phase, the operating characteristics such as internal impedance changes, control response speed, and charging acceptance capability can be reflected, providing a data foundation for energy storage module consistency analysis and performance evaluation.

[0043] Statistical analysis was performed on the current ramp-up process records for each energy storage module. All current sampling values ​​for each module from the start of charging to the stable charging phase were extracted, and a charging current sequence was constructed according to time order. The mean of this sequence was then calculated to obtain the average charging current for the corresponding energy storage module, which characterizes the overall charging capability of the module during the ramp-up phase. After calculating the mean for each module, the charging mean values ​​of all energy storage modules were summarized and statistically analyzed to calculate the overall average reference value for the current charging task. Then, the degree of difference between the charging mean of each energy storage module and the overall average reference value was calculated to obtain the corresponding mean deviation results. The smaller the mean deviation, the higher the consistency between the module and the overall operating state; the larger the mean deviation, the more likely the module has increased internal impedance, accelerated capacity decay, or abnormal control response. To improve the stability of the analysis results, the mean deviation results of multiple charging cycles can be cumulatively statistically analyzed to form long-term consistency evaluation data. The average charging deviation value of each energy storage module is obtained, and the association with the unique identification code of the equipment is established, thereby forming the consistency evaluation result of the energy storage module, which provides a quantitative basis for subsequent energy storage module status diagnosis, equalization control strategy adjustment and maintenance decision-making.

[0044] In one embodiment, step S5 specifically involves the following steps: Based on the average charging deviation value, perform active equalization adjustment of the bidirectional DC / DC converter to obtain the module equalization adjustment result; Based on the results of module balance adjustment, the actual available capacity of the modules is calculated, and a list of actual available capacities is generated. The upper limit of charging and discharging power is calculated based on the actual available capacity list to obtain the upper limit of power carrying capacity for each module.

[0045] In this embodiment, active balancing control is performed based on the average charging deviation value of each energy storage module. When the average charging deviation value of a certain energy storage module exceeds a preset balancing threshold, it is determined that there is a difference in charging capacity or an uneven energy distribution between this module and other modules. At this time, the active balancing mechanism is activated through the corresponding bidirectional DC / DC converter to dynamically adjust the energy flow between modules. During the adjustment process, parameters such as the estimated SOC value, predicted SOH value, battery terminal voltage, and current charging current of the target module are read. The SOC and SOH values ​​are maintained in the range of 0% to 100%, the voltage sampling range is maintained in the range of 0 V to 1000 V, and the current sampling range is maintained in the range of 0 A to 500 A. For modules with a high SOC or an average charging value significantly higher than the overall average level, their charging current allocation ratio is appropriately reduced; for modules with a low SOC or an average charging value lower than the overall average level, their charging current allocation ratio is appropriately increased. The bidirectional DC / DC converter changes the power conversion duty cycle in real time according to the adjustment command to realize energy transfer and power redistribution between modules. During the adjustment period, the current changes of each module are continuously monitored, and the mean deviation results are updated in real time at a sampling frequency of 1 kHz. When the mean deviation within multiple consecutive sampling windows decreases to within a preset allowable range, the equalization adjustment is considered complete. Module equalization adjustment results are generated, including the adjusted SOC state, the change in equalization current, and the redistribution of module energy, thereby improving the consistency level between energy storage modules and reducing the risk of overcharging, over-discharging, and accelerated aging of local modules.

[0046] Based on the operational status after equalization adjustment, the actual usable capacity of each energy storage module is evaluated and calculated. Parameters such as SOC status, SOH prediction results, battery terminal voltage, cumulative charge / discharge capacity, and cycle count are read from the equalization adjustment results, with SOC and SOH values ​​maintained between 0% and 100%. Subsequently, an actual usable capacity evaluation model is established based on the module's rated capacity information. During the evaluation process, the rated capacity is no longer directly used as the dispatchable capacity; instead, the calculation is corrected by comprehensively considering battery aging, the current state of charge, and the operational status after equalization. For example, when a module's SOH decreases, its actual usable capacity is correspondingly lower than its rated capacity; when the SOC approaches the upper or lower limit, the effective capacity that can participate in dispatch also decreases simultaneously. Simultaneously, based on the energy distribution after equalization adjustment, the release capacity and absorbable capacity of each module are evaluated separately, calculating the capacity space that can safely participate in energy dispatch under the current operating conditions. To ensure the reliability of the capacity evaluation results, verification analysis is also performed based on the capacity change trends over several recent operating cycles to avoid evaluation errors caused by fluctuations in a single operational state. The actual available capacity of each energy storage module is obtained, and a correlation is established according to the unique identification code of the equipment to form a list of actual available capacities. This capacity list can accurately reflect the effective energy storage capacity of each energy storage module at the current stage, providing an accurate basis for power scheduling and operation control.

[0047] The charging and discharging capabilities of each energy storage module are dynamically evaluated based on the actual available capacity list, and the corresponding power load limit is calculated. Parameters such as actual available capacity, estimated SOC, predicted SOH, rated power, and current operating status are read, with the rated power range maintained between 100 W and 100 kW. Then, based on the current remaining available capacity and allowable operating range of the energy storage module, its maximum charging power and maximum discharging power are calculated. When the SOC approaches 100%, the charging power limit is reduced to avoid overcharging risk; when the SOC approaches 0%, the discharging power limit is reduced to avoid over-discharging. Simultaneously, the power capacity is corrected based on the SOH prediction results. Modules with higher health status are allowed to undertake a higher proportion of power output tasks, while modules with lower SOH have their power load appropriately reduced to slow down capacity decay. The calculation process also comprehensively considers constraints such as bus voltage fluctuations, battery temperature status, and the rated capacity of the bidirectional DC / DC converter, and makes a safety correction to the theoretical power limit. After the calculations are completed, the maximum charging power limit and maximum discharging power limit for each energy storage module are generated, and a power carrying capacity mapping relationship is established to obtain the power carrying capacity limit for each module. Through dynamic power boundary management, the energy storage system can rationally allocate load tasks according to the actual operating status of the modules, improve the overall energy storage resource utilization rate, and enhance the safety, stability, and economy of modular DC microgrid operation.

[0048] In one embodiment, step S6 specifically involves the following steps: Based on the power carrying capacity limit of each module, the charging and discharging power is predicted to obtain the power prediction value of different modules; The liquid cooling power requirement is calculated based on the predicted power value to obtain the liquid cooling power requirement for different modules; Intelligent cooling parameter adjustment and cooling control processing are performed based on the liquid cooling power requirements.

[0049] In this embodiment, the charging and discharging power during future operating cycles is predicted and analyzed based on the power carrying capacity limit corresponding to each energy storage module. Parameters such as the maximum charging power limit, maximum discharging power limit, estimated SOC, predicted SOH, and actual available capacity of each module are read, with SOC and SOH values ​​ranging from 0% to 100%, and rated power ranging from 100 W to 100 kW. Subsequently, a power prediction model is established by combining the current bus load demand, photovoltaic power generation changes, and historical power operation curves to analyze the power change trend over a future period. During the prediction process, time series data is used as the basic input, and the future charging and discharging demands of the energy storage modules are calculated on a rolling basis to obtain the target power values ​​corresponding to different times. Modules with higher SOH and larger actual available capacity are assigned a higher power carrying weight; modules with lower SOH or smaller actual available capacity are appropriately reduced in power carrying ratio. By continuously updating the load demand and energy storage status parameters, the power prediction values ​​for each module's future operating phase are formed. The predicted power values ​​can reflect the power change trends and load-bearing capacity of different energy storage modules in the future operating cycle, providing a power basis for thermal management resource allocation and liquid cooling system regulation.

[0050] The potential heat load during operation of each energy storage module is calculated based on its predicted power value, and the liquid cooling power requirement is assessed accordingly. Parameters such as the predicted power value, battery terminal voltage, charging / discharging current, battery temperature, and SOH prediction results for each module are read, and the corresponding heat generation is calculated based on the energy loss characteristics during module operation. Since higher energy storage module power results in greater heat generation from internal electrochemical reactions and conductor losses, the predicted power value is used as a crucial basis for heat assessment. Simultaneously, a heat load analysis model is established by combining the battery temperature change rate and historical temperature rise patterns to estimate the temperature rise trend during future operating cycles, forming a set of corresponding module temperature rise prediction curves. The required heat dissipation capacity for each module is calculated based on the prediction results and converted into the corresponding liquid cooling power requirement. For modules expected to experience rapid temperature rise or continuous high-load operation, the liquid cooling requirement level is increased; for modules operating under low load, the liquid cooling resource allocation ratio is reduced. The liquid cooling power requirement results for different energy storage modules are obtained, and a liquid cooling thermal management requirement list is formed, providing a control basis for the coordinated adjustment of liquid cooling and air-cooling equipment.

[0051] Intelligent cooling control is implemented based on the liquid cooling power demand. The system reads the liquid cooling power demand values ​​for each energy storage module and the module temperature rise prediction curve set to analyze future thermal state trends. When the prediction indicates a significant temperature rise risk for a certain energy storage module within the next 5 to 10 minutes, the liquid cooling pre-cooling mode is activated in advance, ensuring the cooling system enters operational status before the temperature rises. A tiered thermal management strategy is used for dynamic adjustment during the control process. When the module temperature is below 30℃, only the air cooling mode is activated, maintaining basic heat dissipation through the fan. When the module temperature is between 30℃ and 35℃, a low-flow liquid cooling mode is activated, combined with air cooling operation, to achieve gentle cooling. When the module temperature exceeds 35℃, the liquid cooling system switches to full-power operation and simultaneously increases the fan speed to improve overall heat dissipation efficiency. When the module temperature exceeds 40℃, both the liquid cooling and air cooling systems operate at full power simultaneously, and a power reduction request is sent to the energy management unit, triggering the EMS power reduction protection mechanism to reduce the charging and discharging load level of the energy storage module and suppress further heat accumulation at the source. During the control process, information such as liquid cooling pump speed, coolant flow rate, fan speed, temperature changes, and power reduction execution status are recorded in real time to form a thermal management response execution record.

[0052] Statistical analysis was performed on the thermal management response execution records to evaluate the temperature convergence effect of each energy storage module during the thermal management control process. Temperature change curves, liquid cooling operating parameters, air cooling operating parameters, and power reduction execution records for each module before and after cooling control were read. The module temperature drop rate, maximum temperature change, and temperature stabilization time were calculated. The temperature convergence status of each module after thermal management intervention was then analyzed. When the module temperature gradually decreased and stabilized within the target temperature range, the thermal management control was deemed effective; when the module temperature continued to fluctuate or still showed a rapid temperature rise trend, it was recorded as a key focus area. During the evaluation process, indicators such as liquid cooling system runtime, fan load, pre-cooling start-up times, and EMS protection trigger times were simultaneously recorded to measure the execution effect of the thermal management strategy and resource consumption. After the evaluation was completed, the temperature convergence results, cooling execution process, temperature control effect, and abnormal temperature rise records of each energy storage module were uniformly archived and linked with the corresponding module's unique identifier to generate an energy storage thermal state management execution record. By accumulating thermal state management data over a long period, reliable data can be provided for thermal safety analysis, cooling strategy optimization, and operation and maintenance decisions of energy storage systems, thereby improving the safety, stability, and service life of modular DC microgrid energy storage systems.

[0053] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0054] In one embodiment, a modular DC microgrid smart energy storage device is provided, which corresponds one-to-one with the methods described in the above embodiments. For example... Figure 5 As shown, the device includes: The acquisition unit is used to acquire module information and obtain the frequency domain power characteristic spectrum when a new module is connected to the DC microgrid. Topology unit, used to perform cross-module correlation analysis based on the frequency domain power characteristic spectrum, and construct microgrid power topology map; The attenuation prediction unit is used to traverse each device based on the power topology map of the microgrid, predict the capacity attenuation of the modules, and output the SOH prediction result set of each module. The deviation calculation unit is used to issue charging instructions to each energy storage module based on the SOH prediction result set, and to calculate the average charging deviation to obtain the average charging deviation value of each module. The upper limit calculation unit is used to calculate the upper limit of charging and discharging power based on the average charging deviation value, so as to obtain the upper limit of power carrying capacity of each module. The cooling control unit is used to calculate the liquid cooling power requirement and control the cooling process based on the power capacity limit of each module.

[0055] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0056] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0057] This application also provides a computer device, such as... Figure 6As shown, the computer device includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor. When the processor executes the computer program, it implements the steps in any of the above method embodiments, or when the processor executes the computer program, it implements the functions of each module / unit in the above device embodiments.

[0058] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete this application. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the computer device.

[0059] Those skilled in the art will understand that Figure 6 The computer device described is merely an example and does not constitute a limitation on the computer device. It may include more or fewer components than shown, or combine certain components, or different components. For example, the computer device may also include input / output devices, network access devices, buses, etc.

[0060] The aforementioned processor can be a Central Processing Unit (CPU), or other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), or Field Programmable Gate Arrays (FPGAs). Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0061] The memory can be an internal storage unit of the computer device, such as a hard drive or RAM. The memory can also be an external storage device of the computer device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the memory can include both internal and external storage units of the computer device.

[0062] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0063] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0064] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0065] In the embodiments provided in this application, it should be understood that the disclosed apparatus / devices and methods can be implemented in other ways. For example, the apparatus / device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0066] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0067] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A modular DC microgrid intelligent energy storage method, characterized in that, Includes the following steps: Step S1: When the new module is connected to the DC microgrid, the module information is collected to obtain the frequency domain power characteristic spectrum; Step S2: Perform cross-module correlation analysis based on the frequency domain power characteristic spectrum to construct a microgrid power topology map; Step S3: Based on the microgrid power topology map, traverse each device one by one, perform module capacity attenuation prediction, and output the SOH prediction result set for each module; Step S4: Based on the SOH prediction result set, issue charging instructions to each energy storage module and calculate the average charging deviation to obtain the average charging deviation value of each module. Step S5: Calculate the upper limit of charging and discharging power based on the average charging deviation value to obtain the upper limit of power carrying capacity of each module; Step S6: Calculate the liquid cooling power requirement and perform cooling control processing based on the upper limit of the power capacity of each module.

2. The modular DC microgrid intelligent energy storage method according to claim 1, characterized in that, The specific steps of step S1 are as follows: When a new module is connected to a DC microgrid, it broadcasts a device description frame via the CAN bus. Handshake confirmation is performed based on the device description frame, and the data is incorporated into the DC microgrid topology management to generate a module registration information table. Electrical parameters are scanned based on the module registration information table to obtain the original voltage and current sampling sequence; The original voltage and current sampling sequence is subjected to abnormal sampling point removal and missing value interpolation repair to obtain a continuous electrical sampling sequence; A short-time Fourier transform is performed on the continuous electrical sampling sequence to obtain the frequency domain power characteristic spectrum.

3. The modular DC microgrid intelligent energy storage method according to claim 2, characterized in that, The device description frame includes device type, rated power, protocol version, and current SOC.

4. The modular DC microgrid intelligent energy storage method according to claim 2, characterized in that, The specific steps of step S2 are as follows: Identify information from multiple modules in a DC microgrid; Based on the frequency domain power characteristic spectrum, cross-module correlation analysis is performed on the information of multiple modules to obtain the power coupling strength coefficient between different modules; Based on the power coupling strength coefficient, topological clustering is performed to obtain the power cooperative domain partitioning result; Communication configuration is performed based on the power coordination domain partitioning results, and the built-in protocol of the adaptation layer is parsed to construct the microgrid power topology diagram.

5. The modular DC microgrid intelligent energy storage method according to claim 4, characterized in that, Step S3 is as follows: Based on the power topology diagram of the microgrid, each device is traversed one by one to extract the operating parameters of each energy storage module; The SOC estimation and extended Kalman filtering are performed on the operating parameters to obtain the SOC estimation value of each module; The estimated SOC value is input into a preset deep learning network to predict module capacity decay and output the SOH prediction result set for each module.

6. The modular DC microgrid intelligent energy storage method according to claim 1, characterized in that, The specific steps of step S4 are as follows: Based on the SOH prediction result set, the energy storage capacity of each module is allocated to generate an energy storage allocation scheme; Based on the energy storage distribution scheme, charging commands are issued to each energy storage module, and charging current information is collected to generate a record of the current ramp-up process. The average charging value of each module is calculated based on the current ramp-up process record, and the mean deviation is calculated to obtain the average charging deviation value of each module.

7. The modular DC microgrid intelligent energy storage method according to claim 6, characterized in that, The specific steps of step S5 are as follows: Based on the average charging deviation value, perform active equalization adjustment of the bidirectional DC / DC converter to obtain the module equalization adjustment result; Based on the results of module balance adjustment, the actual available capacity of the modules is calculated, and a list of actual available capacities is generated. The upper limit of charging and discharging power is calculated based on the actual available capacity list to obtain the upper limit of power carrying capacity for each module.

8. The modular DC microgrid intelligent energy storage method according to claim 7, characterized in that, The specific steps of step S6 are as follows: Based on the power carrying capacity limit of each module, the charging and discharging power is predicted to obtain the power prediction value of different modules; The liquid cooling power requirement is calculated based on the predicted power value to obtain the liquid cooling power requirement for different modules; Intelligent cooling parameter adjustment and cooling control processing are performed based on the liquid cooling power requirements.

9. A modular DC microgrid intelligent energy storage device, characterized in that, The steps for implementing the modular DC microgrid smart energy storage method as described in any one of claims 1 to 8 include: The acquisition unit is used to acquire module information and obtain the frequency domain power characteristic spectrum when a new module is connected to the DC microgrid. Topology unit, used to perform cross-module correlation analysis based on the frequency domain power characteristic spectrum, and construct microgrid power topology map; The attenuation prediction unit is used to traverse each device based on the power topology map of the microgrid, predict the capacity attenuation of the modules, and output the SOH prediction result set of each module. The deviation calculation unit is used to issue charging instructions to each energy storage module based on the SOH prediction result set, and to calculate the average charging deviation to obtain the average charging deviation value of each module. The upper limit calculation unit is used to calculate the upper limit of charging and discharging power based on the average charging deviation value, so as to obtain the upper limit of power carrying capacity of each module. The cooling control unit is used to calculate the liquid cooling power requirement and control the cooling process based on the power capacity limit of each module.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the modular DC microgrid smart energy storage method as described in any one of claims 1 to 8.

Images