Micro-grid energy storage system convenient for energy management connection

Abstract

The invention discloses a micro-grid energy storage system convenient for energy management connection, and relates to the technical field of micro-grid energy storage management, and the system comprises a data collection module which collects a multi-node real-time voltage waveform; the feature analysis module generates a power grid state feature vector containing a harmonic distortion index, a voltage fluctuation depth and a system frequency deviation rate through a feature extraction network; the strategy matching module calls an energy storage control strategy library to generate a charging and discharging instruction set; the power control module executes the instruction and collects internal state parameters such as single battery voltage, battery cluster current and energy storage system temperature; and the health assessment module associates the internal parameters with the power grid feature vectors to generate health degree assessment, and dynamically adjusts the matching weight of the strategy library. According to the system, the power grid state feature precision is improved through the feature extraction network, strategy library dynamic optimization is realized through real-time association of internal states and power grid features, and the adaptability and reliability of energy management connection of the energy storage system are enhanced.

Description

Technical Field

[0001] This invention belongs to the field of microgrid energy storage management technology, specifically a microgrid energy storage system that is easy to connect for energy management. Background Technology

[0002] Microgrid energy storage systems require energy management connections to achieve efficient operation. Existing technologies collect data through voltage sensors, extract features such as harmonics and fluctuations using manually designed algorithms such as Fourier transform and sliding window statistics, and combine this with a pre-set strategy library to control the charging and discharging of energy storage, while independently monitoring battery voltage, current, temperature, and other states. However, traditional feature extraction relies on fixed mathematical models, making it difficult to capture the high-dimensional dynamic information of complex multi-node voltage waveforms, resulting in significant deviations between feature representations and actual grid conditions. Furthermore, the separation of energy storage system health assessment and strategy control makes it impossible to adjust strategy priorities in real time based on internal state changes, easily leading to performance degradation due to mismatch between strategies and the actual battery tolerance.

[0003] Existing solutions fail to address two key issues: first, how to overcome the limitations of manual feature extraction and automatically generate more accurate quantitative features reflecting the grid status from real-time voltage waveforms at multiple locations; and second, how to achieve real-time correlation between internal energy storage state parameters and external grid characteristics, allowing the strategy library to dynamically optimize weights based on battery health, thus avoiding insufficient adaptability of static strategies under complex operating conditions. Summary of the Invention

[0004] This invention aims to solve at least one of the technical problems existing in the prior art; Therefore, this invention proposes a microgrid energy storage system that facilitates energy management connections, comprising: The data acquisition module collects real-time voltage waveform data of nodes at different access locations in the microgrid through a multi-point voltage sampling device; The feature analysis module is used to input the real-time voltage waveform data into the feature extraction network for analysis and processing, and generate a power grid state feature vector. The power grid state feature vector includes harmonic distortion index, voltage fluctuation depth and system frequency deviation rate. The strategy matching module is used to call a preset energy storage control strategy library for matching and querying based on the power grid state feature vector, and generate a charging and discharging instruction set for the energy storage unit. The power control module is used to control the power switching devices of the energy storage converter according to the charging and discharging instruction set, perform energy throughput operations on the energy storage unit, and collect the internal state parameters of the energy storage unit during the execution process. The internal state parameters include the individual cell voltage, the battery cluster current, and the energy storage system temperature. The health assessment module is used to perform real-time correlation analysis between the internal state parameters and the power grid state feature vector to generate a health assessment result of the energy storage system, and dynamically adjust the matching weight of the energy storage control strategy library based on the health assessment result.

[0005] Furthermore, the real-time voltage waveform data is input into a feature extraction network for analysis and processing to generate a power grid state feature vector, including: The real-time voltage waveform data includes a continuous sequence of changes in voltage amplitude, phase angle, and frequency; The continuous variation sequence of voltage amplitude is normalized by a sliding window to eliminate baseline drift caused by the measuring device, thereby obtaining a normalized voltage amplitude sequence. The adjacent period difference calculation is performed on the continuous change sequence of the phase angle to extract the timestamp and jump amplitude of the phase jump event and form a phase stability index. The frequency variation sequence is subjected to spectral decomposition to separate the fundamental component and harmonic components within a preset order range, and the ratio of the amplitude of each harmonic component to the amplitude of the fundamental component is calculated. Based on the standardized voltage amplitude sequence, the percentage of the difference between the maximum and minimum voltage values ​​within each sliding window relative to the rated voltage is calculated as the voltage fluctuation depth; The phase stability index, the ratio of each harmonic component, and the voltage fluctuation depth are combined according to a preset feature encoding rule to form a multidimensional power grid state feature vector.

[0006] Furthermore, based on the power grid state feature vector, a preset energy storage control strategy library is invoked for matching and querying to generate a charging and discharging instruction set for the energy storage unit, including: The charge / discharge instruction set includes the target charge / discharge power, the target power change rate, and the allowed operating time window; The similarity of each feature dimension in the power grid state feature vector with the pre-stored strategy triggering conditions in the energy storage control strategy library is calculated, and the strategy triggering conditions correspond to the historical operation scenarios of the power grid. The energy storage control strategies corresponding to the strategy triggering conditions whose similarity calculation values ​​exceed the matching threshold are selected as the candidate control strategy set; For each energy storage control strategy in the candidate control strategy set, its preset desired control target is extracted. The desired control target includes the desired voltage regulation amount, the desired frequency recovery speed, and the desired harmonic suppression effect. Based on the current dispatchable capacity of the energy storage unit and the battery cluster current and system temperature in the internal state parameters, calculate the power output capability and safety margin of the energy storage unit at the current moment. Combining the desired control objective with the power output capability and safety margin, the target charge / discharge power, the target power change rate, and the allowable operating time window are generated through multi-objective optimization calculations.

[0007] Furthermore, the power switching devices of the energy storage converter are controlled according to the charging and discharging command set to perform energy throughput operations on the energy storage unit, and the internal state parameters of the energy storage unit are collected during the execution process, including: The target charging and discharging power and the target power change rate are converted into duty cycle change curves of the pulse width modulation signals of each bridge arm in the energy storage converter. The power switching device is driven to periodically turn on and off according to the duty cycle change curve, thereby controlling the direction and magnitude of the current flowing through the energy storage unit port; Within the allowed operating time window, the potential difference across each individual cell in the energy storage unit is simultaneously measured at a sampling frequency higher than the power frequency to obtain the real-time distribution of the individual cell voltage. Measure the total current flowing into or out of the energy storage unit battery cluster to obtain a continuous waveform of the battery cluster current; The real-time temperature of the energy storage system is read by temperature sensors placed at key hot spots in the energy storage unit.

[0008] Furthermore, the internal state parameters are correlated with the grid state feature vector in real time to generate a health assessment result for the energy storage system. Based on this health assessment result, the matching weights of the energy storage control strategy library are dynamically adjusted, including: The real-time distribution of the individual cell voltage is statistically analyzed to calculate the voltage consistency index and the slope of the voltage decay trend. The continuous waveform of the battery cluster current is integrated to calculate the cumulative charge throughput, and the current capacity retention rate of the energy storage unit is estimated by combining historical cumulative data. The effectiveness of the heat dissipation system is evaluated by analyzing the real-time temperature values ​​of the energy storage system and the rate of increase of the temperature within the allowable operating time window. The voltage consistency index, voltage decay trend slope, current capacity retention rate, and heat dissipation system performance evaluation results are correlated and mapped with the current grid state feature vector to find the degradation mode of energy storage unit performance under similar grid operating conditions. Based on the degree of deviation between the found degradation patterns and the preset health benchmark, a quantitative health assessment result of the energy storage system is generated. Based on the health assessment results of the energy storage system, the matching weights of strategies involving charge / discharge depth and power change rate in the energy storage control strategy library are adjusted so that in subsequent matching queries, the priority of strategies whose energy storage unit operating load exceeds a preset safety threshold is reduced.

[0009] Furthermore, it also includes: Establish a standard communication connection with the microgrid central energy management system to receive global scheduling instructions and microgrid topology change information from the central energy management system; The global scheduling instructions are analyzed to extract their power demand and planning curves for the energy storage system; The planned curve is coordinated and verified with the charging and discharging instruction set generated based on the local real-time status. When there is a conflict, the final coordinated charging and discharging instruction is generated based on the preset priority rules. The microgrid topology change information is updated to the local topology model, which is used to determine the effectiveness of the multi-point voltage sampling device and the weight of the sampling data. The locally generated health assessment results and key operation logs of the energy storage system are reported to the central energy management system via the standard communication connection.

[0010] Furthermore, the planned curve is coordinated and verified with the charging and discharging instruction set generated based on the local real-time status. When conflicts exist, a final coordinated charging and discharging instruction is generated based on a preset priority rule, including: Compare the power demand in the same time period in the planned curve with the target charge and discharge power concentrated in the same time period of the charge and discharge command; If the sign of the power demand is the same as the sign of the target charge / discharge power but the difference in value is less than the preset tolerance, then the target charge / discharge power is updated based on the power demand of the planned curve. If the sign of the power demand is opposite to the sign of the target charge / discharge power, it is determined to be an instruction conflict; In the event of a command conflict, arbitration is carried out according to a preset priority rule, which specifies the priority order among three needs: ensuring the safe and stable operation of the microgrid, executing central dispatch commands, and maintaining the health of the energy storage system itself. Based on the arbitration result, a decision is made on whether to use the planned curve, the locally generated charge / discharge instruction set, or a power value obtained by weighting the power requirement based on the planned curve and the target charge / discharge power, thereby forming the coordinated charge / discharge instruction.

[0011] Furthermore, it also includes preventative maintenance management of energy storage units, as detailed below: The internal state parameters and the corresponding power grid state feature vector are continuously recorded for each energy throughput operation to form a historical operation database. From the historical operation database, the operation records corresponding to the power grid state feature vectors with high harmonic distortion index and large system frequency deviation rate are selected; By analyzing the fluctuations in the voltage of the individual cells and the peak values ​​of the temperature of the energy storage system in the operation records, characteristic operating conditions that generate stress on the electrochemical system of the energy storage unit are identified. For each identified characteristic operating condition mode, a cumulative stress coefficient is preset, and in subsequent operation, whenever the power grid state is detected to be close to the characteristic operating condition mode, the corresponding cumulative stress coefficient is accumulated. When the cumulative stress coefficient corresponding to a certain characteristic operating condition exceeds its preset maintenance threshold, a preventive maintenance warning is generated for the energy storage unit, and it is recommended to adjust the control strategy to avoid the characteristic operating condition.

[0012] Furthermore, by analyzing the fluctuations in the voltage of the individual battery cells and the peak temperatures of the energy storage system in the operation records, characteristic operating conditions that generate stress on the electrochemical system of the energy storage unit are identified, including: For the selected operation records, extract the maximum voltage value, minimum voltage value, and maximum voltage change rate of the individual battery during the charging and discharging process; Extract the initial temperature, maximum temperature, and time taken for the temperature to reach the peak value of the energy storage system during operation. The maximum voltage value, the minimum voltage value, the maximum voltage change rate, the starting temperature, the highest temperature, and the time taken for the temperature to reach its peak value are compared with the safety boundary values ​​in the energy storage unit's technical specifications. The operating conditions defined by the corresponding power grid state feature vector of those operating records where one or more parameters are close to or repeatedly touch the safety boundary value are marked as the characteristic operating condition mode. A feature label is established for each of the aforementioned characteristic operating conditions, and the feature label contains the range of major power grid characteristic parameters that cause stress.

[0013] Furthermore, it also includes adaptive impedance identification of the grid connection point of the energy storage system, as detailed below: At the grid connection point of the energy storage system, a set of tiny current disturbance signals with a specific spectrum are injected; The voltage response signal at the grid connection point is measured synchronously during the injection of the small current disturbance signal; Based on the injected current disturbance signal and the measured voltage response signal, the equivalent impedance amplitude-frequency characteristic and phase-frequency characteristic viewed from the energy storage system side to the grid side are calculated. Based on the calculated equivalent impedance amplitude-frequency characteristics and phase-frequency characteristics, the equivalent short-circuit capacity and main resonant frequency points of the power grid are identified. The identified equivalent short-circuit capacity and main resonant frequency points are used as input parameters to assist in generating the power grid state feature vector or adjusting the strategy parameters in the energy storage control strategy library.

[0014] Compared with the prior art, the beneficial effects of the present invention are: A feature extraction network is employed to perform end-to-end analysis of real-time voltage waveform data collected by multi-point voltage sampling devices at different access locations, directly generating a grid state feature vector containing harmonic distortion index, voltage fluctuation depth, and system frequency deviation rate. Unlike traditional manually designed algorithms that rely on Fourier transform and sliding window statistics, this network can autonomously learn the complex temporal and frequency domain relationships of voltage waveforms, transforming the high-dimensional raw waveforms into structured features. This enhances the ability to capture transient and steady-state anomalies in the grid, making the feature vector more realistically reflect the collaborative state of multiple nodes and providing input data that closely aligns with actual operating conditions for strategy matching.

[0015] During charging and discharging, the power control module collects internal state parameters such as individual battery voltage, battery cluster current, and energy storage system temperature. These parameters are then correlated with the grid state feature vector generated by the feature analysis module in real time to generate a health assessment result and dynamically adjust the matching weights of the energy storage control strategy library. This changes the conventional model where external grid control and internal battery health operate independently. When abnormal battery cluster current or high temperature is detected, the matching priority of high-risk charging and discharging strategies is automatically reduced, ensuring that strategy selection is adapted to the current battery tolerance capacity. This avoids inappropriate energy throughput from exacerbating system losses, forming a closed-loop adaptive mechanism encompassing grid state, strategy execution, internal feedback, and strategy optimization. Attached Figure Description

[0016] Figure 1 This is a timing diagram of the microgrid energy storage system for easy energy management connection as described in this invention; Figure 2 A flowchart for matching energy storage control strategies and generating charge / discharge commands; Figure 3 A diagram showing the power command coordination verification results for a microgrid energy storage system; Figure 4 This is a diagram showing the health assessment results of a microgrid energy storage system. Figure 5 This is a comparison chart of the power grid impedance characteristics under different operating conditions. Detailed Implementation

[0017] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. 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.

[0018] See Figure 1 The data acquisition module first collects real-time voltage waveform data from nodes at different access points in the microgrid using multi-point voltage sampling devices. Subsequently, the feature analysis module inputs the collected real-time voltage waveform data into a built-in feature extraction network for analysis. This network processes the waveform data and outputs a grid state feature vector, which contains multiple dimensions characterizing the grid's power quality, specifically harmonic distortion index, voltage fluctuation depth, and system frequency deviation rate. Upon receiving the grid state feature vector, the strategy matching module uses it as a query condition to call a locally pre-set energy storage control strategy library for matching. The strategy library pre-stores control strategies corresponding to different grid states. Through matching, a set of charging and discharging command sets for the energy storage unit under the current grid state is generated. The power control module is responsible for executing this charging and discharging command set. It parses the commands into specific control signals to drive the power switching devices in the energy storage converter, thereby controlling the energy throughput operation of the energy storage unit. While executing the energy throughput operation, the power control module simultaneously collects the internal state parameters of the energy storage unit, including individual cell voltage, cell cluster current, and energy storage system temperature. The health assessment module performs real-time correlation analysis between the internal state parameters collected by the power control module and the grid state feature vector generated by the feature analysis module. By analyzing the correlation between them, a quantitative health assessment result of the energy storage system is generated. The module also dynamically adjusts the matching weight of each strategy in the energy storage control strategy library based on this health assessment result, so that the strategy matching is more inclined to select the control strategy that is beneficial to the health of the energy storage system.

[0019] See Figure 2In one embodiment of the present invention, the process of the feature extraction network processing real-time voltage waveform data involves multiple steps. The real-time voltage waveform data comes from multi-point voltage sampling devices at different access points in a microgrid. In an example scenario, voltage distortion occurs at the microgrid access point due to nonlinear load switching. The voltage amplitude sequence shows a periodic distorted waveform, the phase angle sequence shows a step change, and the frequency sequence exhibits low-frequency fluctuations. The feature extraction network performs sliding window normalization on the continuous voltage amplitude variation sequence. The sliding window width is set to 20 power frequency cycles. The normalization process is achieved by removing the linear trend within the window and normalizing to zero mean, eliminating baseline drift caused by measurement device gain error, and obtaining a normalized voltage amplitude sequence. Data comparison shows that the voltage amplitude sequence before normalization has a slow upward trend, while the trend of the sequence after normalization is eliminated and the amplitude fluctuations are clearer. At the same time, the adjacent cycle difference calculation is performed on the continuous phase angle variation sequence. The difference calculation calculates the difference of the phase value at the same electrical angle point in each power frequency cycle to extract the phase. The timestamps and amplitudes of transition events are used to form a phase stability index composed of the number of transition events and the average amplitude. A spectral decomposition is performed on the continuous frequency variation sequence. The spectral decomposition uses a windowed Fourier transform to obtain the fundamental component and the 2nd to 13th harmonic components. The ratio of the amplitude of each harmonic component to the amplitude of the fundamental component is calculated to obtain the components of the harmonic distortion index. Based on the standardized voltage amplitude sequence, the difference between the maximum and minimum voltage values ​​within each sliding window is calculated. The percentage obtained by comparing the difference with the system's rated voltage is used as the voltage fluctuation depth. In a data comparison, the rated voltage is 380 volts, the maximum voltage value within the window is 385 volts, and the minimum voltage value is 375 volts, resulting in a voltage fluctuation depth of 2.63%. Finally, the feature extraction network integrates the phase stability index, the ratios of each harmonic component, and the voltage fluctuation depth, and combines them into a power grid state feature vector according to a preset feature encoding rule. The feature encoding rule uses the phase stability index as the first dimension, the harmonic ratios as subsequent dimensions in sequence, and the voltage fluctuation depth as the last dimension.

[0020] In some embodiments, the process of generating a charging and discharging instruction set by calling a preset energy storage control strategy library based on the grid state feature vector is as follows: The charging and discharging instruction set includes the target charging and discharging power, the target power change rate, and the allowed operating time window. The matching query calculates the similarity between each feature dimension in the grid state feature vector and the pre-stored strategy triggering conditions in the energy storage control strategy library. The strategy triggering conditions correspond to historical grid operation scenarios, including photovoltaic power output surges and load step jumps. The similarity calculation uses the cosine similarity method, and the formula is: in: This represents the calculated similarity value between the power grid state feature vector and the triggering condition of the q-th strategy. This represents the value of the r-th feature dimension of the power grid state feature vector. Let represent the r-th feature dimension value of the q-th strategy trigger condition, and m be the total number of feature dimensions. Energy storage control strategies corresponding to trigger conditions whose similarity calculation values ​​exceed the matching threshold are selected as the candidate control strategy set. The matching threshold is set to 0.85. For each energy storage control strategy in the candidate control strategy set, its preset expected control objective is extracted. The expected control objective includes the expected voltage regulation, the expected frequency recovery rate, and the expected harmonic suppression effect. For example, a strategy might expect to recover the frequency deviation to within 0.1 Hz within 0.2 seconds. Based on the current schedulable capacity of the energy storage unit and the battery cluster current and system temperature in the internal state parameters, the power output capability and safety margin of the energy storage unit at the current moment are calculated. The power output capability is determined based on the battery cluster current limit and the system temperature derating curve. The safety margin is calculated based on the ratio of schedulable capacity to the maximum allowable capacity. Combining the expected control objective with the power output capability and safety margin, target charge / discharge power, target power change rate, and allowable operating time window are generated through multi-objective optimization calculation. Multi-objective optimization uses linear programming to solve for the optimal solution that satisfies the constraints.

[0021] In one embodiment of the present invention, the power control module controls the energy storage converter and collects internal state parameters according to the charge and discharge command set. The charge and discharge command set includes the target charge and discharge power, the target power change rate, and the allowable operating time window. In an example scenario, the charge and discharge command set requires a target charge and discharge power of 100 kW (positive), a target power change rate of 50 kW / s, and an allowable operating time window of the next 5 minutes. The power control module converts the target charge and discharge power and the target power change rate into the duty cycle change curve of the pulse width modulation signal of each arm in the energy storage converter. The conversion is based on the DC bus voltage and the converter topology. The target charge and discharge power of 100 kW corresponds to a steady-state duty cycle, and the target power change rate of 50 kW / s corresponds to the linear change rate of the duty cycle within a specific time period. Data comparison shows that when the target power change rate increases from 10 kW / s to 50 kW / s, the slope of the duty cycle change curve increases by 4 times. The power control module drives the power switching device to periodically turn on and off according to the duty cycle change curve, thereby controlling the flow through the energy storage unit port. The direction and magnitude of the current are controlled within an example scenario. During the permitted operating time window, the power switching devices operate at a switching frequency of 10 kHz. The port current linearly increases from 0 Amperes to 200 Amperes within 2 seconds of the command being issued. Within this permitted operating time window, the power control module synchronously measures the potential difference across each individual cell in the energy storage unit at a sampling frequency higher than the power frequency. The sampling frequency is set to 10 kHz to obtain the real-time distribution of the individual cell voltages. Data comparison shows the differences in voltage distribution among different individual cells within the same battery cluster during charging. The highest voltage cell is 3.65 V, and the lowest is 3.55 V. The power control module measures the total current flowing into or out of the energy storage unit's battery cluster, obtaining a continuous waveform of the battery cluster current. In an example scenario, the battery cluster current exhibits a linearly increasing waveform in the early stages of the permitted operating time window and a constant waveform in the later stages. The power control module reads the real-time temperature of the energy storage system through temperature sensors located at key hot spots in the energy storage unit. These key hot spots include the middle of the battery cluster and the inverter heatsink. The temperature sensors read once per second.

[0022] In some embodiments, the health assessment module performs real-time correlation analysis between internal state parameters and grid state feature vectors to generate a health assessment result for the energy storage system. Based on this result, it dynamically adjusts the matching weights of the energy storage control strategy library. The module also performs statistical analysis on the real-time distribution of individual battery voltages to calculate voltage consistency indices and voltage decay trend slopes. The voltage consistency indices are represented by the standard deviation of the individual battery voltages; in one example scenario, the standard deviation of the real-time distribution is 0.03 volts. The voltage decay trend slope is obtained by linearly fitting the peak values ​​of individual battery voltages over multiple charge-discharge cycles. Furthermore, the module integrates the continuous waveform of the battery cluster current to calculate the cumulative charge throughput and estimates the current capacity retention rate of the energy storage unit by combining historical cumulative data. The cumulative charge throughput is obtained by integrating the battery cluster current as a function of time. The current capacity retention rate is the ratio of the current cumulative charge throughput to the initial nominal capacity. The real-time temperature values ​​of the energy storage system and its rise slope within the allowable operating time window are analyzed to evaluate the effectiveness of the cooling system. The temperature rise slope is the ratio of temperature change to time. The cooling system effectiveness evaluation result is characterized by the ratio of the temperature rise rate to the thermal design specification. The health assessment module correlates and maps voltage consistency indicators, voltage decay trend slope, current capacity retention rate, and cooling system effectiveness evaluation results with the current grid state characteristic vector to find degradation modes of energy storage unit performance under similar grid operating conditions. The correlation mapping is achieved by querying the historical operating database, which records the long-term variation trends of internal state parameters under different grid state characteristic vectors. Based on the deviation of the found degradation mode from the preset health benchmark, a quantitative energy storage system health assessment result is generated. The health benchmark is the performance parameters of the energy storage unit in a completely new state. The health assessment result is a multi-dimensional distance measure between the current parameters and the health benchmark. An example formula is: in: This indicates the health assessment results. This represents the weight of the j-th health parameter. This represents the value of the j-th health parameter. Let represent the baseline health value corresponding to the j-th health parameter, and t represent the total number of health parameters. Based on the health assessment results of the energy storage system, the matching weights of strategies involving charge / discharge depth and power change rate in the energy storage control strategy library are adjusted so that the priority of strategies whose energy storage unit operating load exceeds the preset safety threshold is reduced in subsequent matching queries. Specifically, when the health assessment result is lower than the threshold, the matching weight of the corresponding strategy is multiplied by an attenuation coefficient less than 1.

[0023] Optionally, the sampling frequency for measuring the voltage of a single cell can be adjusted according to the cell type; a higher sampling frequency can be used for cells with high dynamic response. It is understood that the temperature sensor can be positioned within the tabs of the cell cluster and inside the housing.

[0024] In one embodiment of the present invention, the system establishes a standard communication connection with the microgrid central energy management system. This standard communication connection uses the IEC61850 protocol. Through this connection, the system receives global dispatch instructions and microgrid topology change information from the microgrid central energy management system. In an example scenario, the global dispatch instructions issued by the microgrid central energy management system are in JSON format, containing the planned power demand curve for the local energy storage system for the next 15 minutes. The microgrid topology change information is transmitted in the form of semaphores, indicating that a certain photovoltaic feeder switch has been disconnected. The system parses the received global dispatch instructions, extracts the power demand and planned curve for the energy storage system from the global dispatch instructions, and then parses... The system identifies the power demand as requiring 50 kW of active power at a specific time point. The planned power curve is a broken line with the timestamp as the horizontal axis and the power value as the vertical axis. The system coordinates and verifies the planned power curve with the charging and discharging command set generated based on the local real-time status. When conflicts exist, the final coordinated charging and discharging command is generated based on preset priority rules. In specific implementation, the coordination and verification includes comparing the power demand in the planned power curve for the same time period with the target charging and discharging power in the charging and discharging command set for the same time period. For example, in the time period from 10:00:00 to 10:05:00, the power demand of the planned power curve is +50 kW (discharging), while the target charging and discharging power of the locally generated charging and discharging command set is... For a power demand of -30 kW (charging), the sign of the power demand is determined to be opposite to the sign of the target charging / discharging power. The system arbitrates according to preset priority rules, which are stored in a text configuration file. These rules stipulate that the demand for ensuring the safe and stable operation of the microgrid has the highest priority, followed by the demand for executing dispatch instructions from the microgrid's central energy management system, and then the demand for maintaining the health of the energy storage system itself. Based on the arbitration result, in the event of a command conflict, if the demand for ensuring the safe and stable operation of the microgrid is triggered, the locally generated charging / discharging command set is used to form a coordinated charging / discharging command; otherwise, the planned curve from the microgrid's central energy management system is used. The system updates the microgrid topology change information to the current configuration file. The local topology model is stored in the form of a node-branch association matrix. Microgrid topology change information is used to update the on / off status of the corresponding branches in the association matrix. The local topology model is used to determine the validity of multi-point voltage sampling devices and the weight of sampling data. For example, when a branch is disconnected, the voltage sampling device at the end of the branch is marked as invalid, and its sampling data has a weight of zero in subsequent analysis. The system reports the locally generated energy storage system health assessment results and key operation logs to the microgrid central energy management system through standard communication connections. The key operation logs include charging and discharging command set execution records and internal status parameter over-limit records. The reports are transmitted in the form of periodic telemetry frames or event triggers.

[0025] In some embodiments, the process of coordinating and verifying the planned curve with the charging and discharging instruction set generated based on local real-time status, and generating the final coordinated charging and discharging instruction based on preset priority rules when conflicts exist, involves the following steps: First, the power demand in the planned curve for the same time period is compared with the target charging and discharging power in the charging and discharging instruction set for the same time period. If the sign of the power demand is the same as the sign of the target charging and discharging power, but the numerical difference is less than a preset tolerance, the target charging and discharging power is updated based on the power demand of the planned curve. The preset tolerance is set to 5% of the rated power. For example, if the rated power is 100 kW, the tolerance is 5 kW. If the sign of the power demand is positive (discharging) and the value is +52 kW, and the sign of the target charging and discharging power is also positive and the value is +48 kW, the difference between the two is 4 kW, which is less than the 5 kW tolerance. Therefore, the +52 kW value is used. The target charging and discharging power is updated. If the sign of the power demand is opposite to the sign of the target charging and discharging power, it is determined to be a command conflict. When a command conflict occurs, arbitration is carried out according to a preset priority rule. The priority rule specifies the order of priority among three needs: ensuring the safe and stable operation of the microgrid, executing central dispatch commands, and maintaining the health of the energy storage system itself. Data comparison shows that when the voltage drops sharply to 0.9 times the rated voltage, the need to ensure the safe and stable operation of the microgrid has the highest priority. The arbitration result determines whether to use the locally generated discharge command with the supporting voltage as the target. Based on the arbitration result, it is determined whether to use the planned curve, the locally generated charging and discharging command set, or the power value obtained by weighting the power demand based on the planned curve and the target charging and discharging power to form a coordinated charging and discharging command. The formula for weighted calculation is: in: This indicates the final power value of the coordinated charge / discharge command. This represents the power demand value of the planned curve. This represents the target charge / discharge power value generated locally. This represents the weighting coefficient determined by the priority arbitration result, when the central dispatch instruction is executed with priority. Set to 1 when prioritizing local control requests. Take 0 when a compromise is needed. Take 0.5.

[0026] Optionally, the standard communication connection for communicating with the microgrid's central energy management system can also use the Modbus TCP protocol. It is understood that microgrid topology change information may also include plug-and-play events for distributed generation sources.

[0027] See Figure 3This diagram illustrates the power command coordination and verification results of a microgrid energy storage system. It shows the process of power command coordination and verification within 15 minutes, focusing on the comparison between the central dispatch plan, locally generated commands, and the final coordination result. During the command conflict phase, the central dispatch requested discharge, but the locally generated command requested charging; the opposite signs led to a conflict. At this point, the priority rule determined that local safety requirements were higher, so the coordinated command adopted the local charging command. During the command convergence phase, the locally generated command switched to discharge, aligning with the central dispatch direction. As time progressed, the coordinated power gradually approached the central dispatch's 50kW target, reflecting the priority shift from local response to executing dispatch commands. In the final convergence phase, the coordinated power approached the central dispatch plan, indicating that in the absence of safety triggering conditions, the system prioritized executing central dispatch commands while also considering energy storage health constraints.

[0028] In one embodiment of the present invention, the system performs preventative maintenance management of the energy storage unit, continuously recording the internal state parameters and corresponding grid state feature vectors for each energy throughput operation, forming a historical operation database. In an example scenario, a photovoltaic power station microgrid performs frequent charging and discharging operations on the energy storage unit during a period of drastic afternoon sunlight fluctuations. The internal state parameters of each operation, such as individual cell voltage, cell cluster current, energy storage system temperature, and corresponding grid state feature vectors, such as harmonic distortion index, voltage fluctuation depth, and system frequency deviation rate, are timestamped and stored in the historical operation database. Operation records corresponding to grid state feature vectors with high harmonic distortion index and large system frequency deviation rate are selected from the historical operation database. The selection criteria are set as a harmonic distortion index greater than 5% and an absolute value of system frequency deviation rate greater than 0.5 Hz. Operation records meeting these criteria are extracted for subsequent analysis. See Table 1.

[0029] Table 1: High-stress operation records selected from the historical operation database Analyzing the fluctuations in individual cell voltages and peak temperatures of the energy storage system in the operation records, characteristic operating conditions that generate stress on the electrochemical system of the energy storage unit were identified. For selected operation records, the maximum, minimum, and maximum voltage values ​​and voltage change rates of individual cells during charge and discharge processes were extracted. For example, from record number 1027, the maximum voltage value was extracted to be 3.70 volts, the minimum voltage value to be 3.48 volts, and the maximum voltage change rate to be 0.15 volts per second. The initial temperature, maximum temperature, and time taken to reach the peak temperature of the energy storage system during operation were also extracted. From the same record, the initial temperature was 26 degrees Celsius, the maximum temperature to be 55 degrees Celsius, and the time taken to reach the peak temperature to be 180 seconds. The maximum, minimum, and voltage change rates were then analyzed. The maximum value, initial temperature, highest temperature, and time taken to reach the peak temperature are compared with the safety boundary values ​​in the energy storage unit's technical specifications. The safety boundary values ​​include the upper limit of the single-cell voltage (3.75 volts), the lower limit of 3.20 volts, the maximum temperature rise rate (5 degrees Celsius per minute), and the maximum allowable temperature (60 degrees Celsius). Operating records with one or more parameters approaching or repeatedly touching the safety boundary values ​​are marked as characteristic operating condition modes according to the corresponding grid state feature vector. For example, in record number 1027, the maximum voltage value of 3.70 volts is close to the upper limit of 3.75 volts, and the highest temperature of 55 degrees Celsius is close to the upper limit of 60 degrees Celsius. The operating condition defined by the corresponding grid state feature vector is marked as a characteristic operating condition mode. A feature label is established for each characteristic operating condition mode. The feature label contains the range of the main grid characteristic parameters that cause stress.

[0030] In some embodiments, a cumulative stress coefficient is preset for each identified characteristic operating condition mode, and the corresponding cumulative stress coefficient is accumulated whenever the grid state is detected to be close to the characteristic operating condition mode during subsequent operation. The initial value of the cumulative stress coefficient is zero. When the current grid state feature vector is detected to match the parameter range defined by the feature label of a certain characteristic operating condition mode, the formula for the cumulative stress coefficient corresponding to that mode is as follows: in: This represents the cumulative stress coefficient updated at time k+1 for the i-th characteristic working condition mode. This represents the cumulative stress coefficient at time k. This represents the preset weighting coefficient for the i-th characteristic working condition mode. The indicator function has a value of 1 if the current grid state matches the i-th mode at time k, otherwise it is 0. When the cumulative stress coefficient corresponding to a certain characteristic operating condition mode exceeds its preset maintenance threshold, a preventive maintenance warning is generated for the energy storage unit, and the control strategy is suggested to be adjusted to avoid the characteristic operating condition mode. For example, the maintenance threshold for the "high harmonics accompanied by high frequency deviation" mode is set to 100. When its cumulative stress coefficient increases from 95 to 110 and exceeds the threshold, the system generates a warning and suggests adjusting the strategy to suppress harmonics or reduce the power response depth under this type of operating condition.

[0031] See Figure 4 This is a health assessment chart of a microgrid energy storage system, showing the trend of its health score from January to June. The core focus is on assessing the evolution of health status across four dimensions: system, battery, temperature, and power. The chart visually presents the multi-dimensional, synergistic changes in the energy storage system's health. System health is a comprehensive reflection of battery, temperature, and power health, and these three are highly correlated. The relative stability of battery health indicates that the aging of the electrochemical system is gradual, while fluctuations in temperature and power have a more direct impact on system health. The continuous downward trend in health provides data support for adjusting preventative maintenance strategies, such as optimizing charging and discharging strategies and strengthening the maintenance of the heat dissipation system during the high-temperature summer months. The linear downward trend of health in each dimension can be used to fit aging models, predict future health changes, and provide a data foundation for the full lifecycle management of energy storage systems.

[0032] In one embodiment of the invention, the system performs adaptive impedance identification at the grid connection point of the energy storage system. At the grid connection point between the energy storage system and the microgrid, a set of small current disturbance signals with a specific spectrum is injected under the control of the energy storage converter. The small current disturbance signals with a specific spectrum are composed of multiple sinusoidal current components of a specific frequency superimposed on each other. These frequency components are typically selected at the subharmonic, interharmonic, and specific harmonic frequencies of the power frequency, such as 25 Hz, 125 Hz, 275 Hz, and 575 Hz components. The amplitude of the injected current is limited to within one to five percent of the rated current to avoid affecting the normal operation of the grid. In an example scenario, during the evening peak load period of the industrial park microgrid... The segment performs adaptive impedance identification. The energy storage converter injects a set of small current disturbance signals containing 35 Hz, 150 Hz and 300 Hz components at the grid connection point. The amplitude of each component is two percent of the rated current. The voltage response signal at the grid connection point is measured simultaneously during the injection of small current disturbance signals of a specific spectrum. The voltage response signal is collected by a high-precision voltage sensor, and the sampling frequency is at least ten times the highest disturbance frequency to ensure signal integrity. Data comparison shows that when a 35 Hz current disturbance is injected, a voltage component of the same frequency appears in the voltage response signal at the grid connection point. The ratio of its amplitude to the amplitude of the injected current reflects the grid impedance amplitude at that frequency point.

[0033] In some embodiments, based on the injected current disturbance signal and the measured voltage response signal, the equivalent impedance amplitude-frequency characteristic and phase-frequency characteristic viewed from the energy storage system side to the grid side are calculated. The calculation process is performed in the frequency domain. A Fast Fourier Transform is performed on the injected small current disturbance signal of a specific spectrum and the measured voltage response signal, respectively, to obtain the current phasor and voltage phasor at each frequency point. The equivalent impedance amplitude-frequency characteristic and phase-frequency characteristic are given by the formula: in: Indicates at frequency point The equivalent complex impedance at that point, Representing frequency point Voltage phasor at the location, Representing frequency point The current phasor at point , the equivalent impedance amplitude-frequency characteristic is With frequency The changing curve, the equivalent impedance phase frequency characteristic is With frequency The changing curves, based on the calculated equivalent impedance amplitude-frequency and phase-frequency characteristics, identify the equivalent short-circuit capacity and main resonant frequency points of the power grid. The equivalent short-circuit capacity is estimated by analyzing the equivalent impedance amplitude at the power frequency. The main resonant frequency points correspond to the frequency points where local maxima appear on the equivalent impedance amplitude-frequency characteristic curve. In a data comparison, the system defaulted to an equivalent short-circuit capacity of 10 MVA before identification, and the actual equivalent short-circuit capacity was calculated to be 8.5 MVA after identification. The main resonant frequency points were identified as 225 Hz and 450 Hz.

[0034] In practical implementation, the identified equivalent short-circuit capacity and main resonant frequency points are used as input parameters to assist in generating the grid state characteristic vector or adjusting the strategy parameters in the energy storage control strategy library. When used to assist in generating the grid state characteristic vector, the main resonant frequency point information can enhance the analysis of grid background harmonics. For example, if a resonant point is identified at 225 Hz, the component near the 4.5th harmonic will be given special attention when the feature analysis module performs spectral decomposition. When used to adjust the strategy parameters in the energy storage control strategy library, the equivalent short-circuit capacity information can be used to correct the control gain related to grid strength in the strategy. For example, under weak grid conditions with a small equivalent short-circuit capacity, the limiting parameters of the power change rate in the strategy will be automatically tightened to prevent oscillations.

[0035] See Figure 5 This is a comparison chart of grid impedance characteristics under different operating conditions. Under all conditions, the peak impedance amplitude appears around 250Hz and 450Hz, indicating that these two frequencies are inherent resonant points of the system, independent of the operating conditions and determined by the grid topology and parameters. At these two resonant frequencies of 250Hz and 450Hz, the impedance amplitude increases significantly, meaning that the system is more prone to harmonic amplification at these frequencies. This provides a clear harmonic suppression target for the control strategy of the energy storage converter, allowing for targeted adjustment of control parameters to avoid generating harmonic currents near the resonant points. The impedance amplitude under short-circuit conditions is much higher than under other conditions, indicating that the equivalent impedance characteristics of the system change drastically under extreme fault scenarios. This requires the energy storage control strategy library to dynamically adjust control parameters under extreme conditions such as short circuits to maintain system stability. The impedance amplitude-frequency curves under different operating conditions have obvious characteristic differences and can be used as feature vectors for operating condition identification. For example, the impedance amplitude reaches 2.5Ω at 250Hz under short-circuit conditions, much higher than under other conditions, which can be used as a diagnostic basis for short-circuit faults.

[0036] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A microgrid energy storage system that is easy to connect for energy management, characterized in that, include: The data acquisition module collects real-time voltage waveform data of nodes at different access locations in the microgrid through a multi-point voltage sampling device; The feature analysis module is used to input the real-time voltage waveform data into the feature extraction network for analysis and processing, and generate a power grid state feature vector. The power grid state feature vector includes harmonic distortion index, voltage fluctuation depth and system frequency deviation rate. The strategy matching module is used to call a preset energy storage control strategy library for matching and querying based on the power grid state feature vector, and generate a charging and discharging instruction set for the energy storage unit. The power control module is used to control the power switching devices of the energy storage converter according to the charging and discharging instruction set, perform energy throughput operations on the energy storage unit, and collect the internal state parameters of the energy storage unit during the execution process. The internal state parameters include the individual cell voltage, the battery cluster current, and the energy storage system temperature. The health assessment module is used to perform real-time correlation analysis between the internal state parameters and the power grid state feature vector to generate a health assessment result of the energy storage system, and dynamically adjust the matching weight of the energy storage control strategy library based on the health assessment result.

2. The microgrid energy storage system for easy energy management connection according to claim 1, characterized in that, The real-time voltage waveform data is input into a feature extraction network for analysis and processing to generate a power grid state feature vector, including: The real-time voltage waveform data includes a continuous sequence of changes in voltage amplitude, phase angle, and frequency; The continuous variation sequence of voltage amplitude is normalized by a sliding window to eliminate baseline drift caused by the measuring device, thereby obtaining a normalized voltage amplitude sequence. The adjacent period difference calculation is performed on the continuous change sequence of the phase angle to extract the timestamp and jump amplitude of the phase jump event and form a phase stability index. The frequency variation sequence is subjected to spectral decomposition to separate the fundamental component and harmonic components within a preset order range, and the ratio of the amplitude of each harmonic component to the amplitude of the fundamental component is calculated. Based on the standardized voltage amplitude sequence, the percentage of the difference between the maximum and minimum voltage values ​​within each sliding window relative to the rated voltage is calculated as the voltage fluctuation depth; The phase stability index, the ratio of each harmonic component, and the voltage fluctuation depth are combined according to a preset feature encoding rule to form a multidimensional power grid state feature vector.

3. The microgrid energy storage system for easy energy management connection according to claim 1, characterized in that, Based on the power grid state feature vector, a preset energy storage control strategy library is invoked for matching and querying to generate a set of charging and discharging instructions for the energy storage unit, including: The charge / discharge instruction set includes the target charge / discharge power, the target power change rate, and the allowed operating time window; The similarity of each feature dimension in the power grid state feature vector with the pre-stored strategy triggering conditions in the energy storage control strategy library is calculated, and the strategy triggering conditions correspond to the historical operation scenarios of the power grid. The energy storage control strategies corresponding to the strategy triggering conditions whose similarity calculation values ​​exceed the matching threshold are selected as the candidate control strategy set; For each energy storage control strategy in the candidate control strategy set, its preset desired control target is extracted. The desired control target includes the desired voltage regulation amount, the desired frequency recovery speed, and the desired harmonic suppression effect. Based on the current dispatchable capacity of the energy storage unit and the battery cluster current and system temperature in the internal state parameters, calculate the power output capability and safety margin of the energy storage unit at the current moment. Combining the desired control objective with the power output capability and safety margin, the target charge / discharge power, the target power change rate, and the allowable operating time window are generated through multi-objective optimization calculations.

4. The microgrid energy storage system for easy energy management connection according to claim 1, characterized in that, The power switching devices of the energy storage converter are controlled according to the charging and discharging command set to perform energy throughput operations on the energy storage unit, and the internal state parameters of the energy storage unit are collected during the execution process, including: The target charging and discharging power and the target power change rate are converted into duty cycle change curves of the pulse width modulation signals of each bridge arm in the energy storage converter. The power switching device is driven to periodically turn on and off according to the duty cycle change curve, thereby controlling the direction and magnitude of the current flowing through the energy storage unit port; Within the allowed operating time window, the potential difference across each individual cell in the energy storage unit is simultaneously measured at a sampling frequency higher than the power frequency to obtain the real-time distribution of the individual cell voltage. Measure the total current flowing into or out of the energy storage unit battery cluster to obtain a continuous waveform of the battery cluster current; The real-time temperature of the energy storage system is read by temperature sensors placed at key hot spots in the energy storage unit.

5. The microgrid energy storage system for easy energy management connection according to claim 4, characterized in that, The internal state parameters are correlated with the grid state feature vector in real time to generate a health assessment result for the energy storage system. Based on this health assessment result, the matching weights of the energy storage control strategy library are dynamically adjusted, including: The real-time distribution of the individual cell voltage is statistically analyzed to calculate the voltage consistency index and the slope of the voltage decay trend. The continuous waveform of the battery cluster current is integrated to calculate the cumulative charge throughput, and the current capacity retention rate of the energy storage unit is estimated by combining historical cumulative data. The effectiveness of the heat dissipation system is evaluated by analyzing the real-time temperature values ​​of the energy storage system and the rate of increase of the temperature within the allowable operating time window. The voltage consistency index, voltage decay trend slope, current capacity retention rate, and heat dissipation system performance evaluation results are correlated and mapped with the current grid state feature vector to find the degradation mode of energy storage unit performance under similar grid operating conditions. Based on the degree of deviation between the found degradation patterns and the preset health benchmark, a quantitative health assessment result of the energy storage system is generated. Based on the health assessment results of the energy storage system, the matching weights of strategies involving charge / discharge depth and power change rate in the energy storage control strategy library are adjusted so that in subsequent matching queries, the priority of strategies whose energy storage unit operating load exceeds a preset safety threshold is reduced.

6. The microgrid energy storage system for easy energy management connection according to claim 1, characterized in that, Also includes: Establish a standard communication connection with the microgrid central energy management system to receive global scheduling instructions and microgrid topology change information from the central energy management system; The global scheduling instructions are analyzed to extract their power demand and planning curves for the energy storage system; The planned curve is coordinated and verified with the charging and discharging instruction set generated based on the local real-time status. When there is a conflict, the final coordinated charging and discharging instruction is generated based on the preset priority rules. The microgrid topology change information is updated to the local topology model, which is used to determine the effectiveness of the multi-point voltage sampling device and the weight of the sampling data. The locally generated health assessment results and key operation logs of the energy storage system are reported to the central energy management system via the standard communication connection.

7. The microgrid energy storage system for easy energy management connection according to claim 6, characterized in that... The planned curve is coordinated and verified with the charging and discharging instruction set generated based on the local real-time status. When conflicts exist, a final coordinated charging and discharging instruction is generated based on preset priority rules, including: Compare the power demand in the same time period in the planned curve with the target charge and discharge power concentrated in the same time period of the charge and discharge command; If the sign of the power demand is the same as the sign of the target charge / discharge power but the difference in value is less than the preset tolerance, then the target charge / discharge power is updated based on the power demand of the planned curve. If the sign of the power demand is opposite to the sign of the target charge / discharge power, it is determined to be an instruction conflict; In the event of a command conflict, arbitration is carried out according to a preset priority rule, which specifies the priority order among three needs: ensuring the safe and stable operation of the microgrid, executing central dispatch commands, and maintaining the health of the energy storage system itself. Based on the arbitration result, a decision is made on whether to use the planned curve, the locally generated charge / discharge instruction set, or a power value obtained by weighting the power requirement based on the planned curve and the target charge / discharge power, thereby forming the coordinated charge / discharge instruction.

8. The microgrid energy storage system for easy energy management connection according to claim 1, characterized in that, This also includes preventative maintenance management of energy storage units, as detailed below: The internal state parameters and the corresponding power grid state feature vector are continuously recorded for each energy throughput operation to form a historical operation database. From the historical operation database, the operation records corresponding to the power grid state feature vectors with high harmonic distortion index and large system frequency deviation rate are selected; By analyzing the fluctuations in the voltage of the individual cells and the peak values ​​of the temperature of the energy storage system in the operation records, characteristic operating conditions that generate stress on the electrochemical system of the energy storage unit are identified. For each identified characteristic operating condition mode, a cumulative stress coefficient is preset, and in subsequent operation, whenever the power grid state is detected to be close to the characteristic operating condition mode, the corresponding cumulative stress coefficient is accumulated. When the cumulative stress coefficient corresponding to a certain characteristic operating condition exceeds its preset maintenance threshold, a preventive maintenance warning is generated for the energy storage unit, and it is recommended to adjust the control strategy to avoid the characteristic operating condition.

9. The microgrid energy storage system for easy energy management connection according to claim 8, characterized in that, The analysis of the operational records, including fluctuations in the voltage of individual cells and peak temperatures of the energy storage system, identifies characteristic operating conditions that stress the electrochemical system of the energy storage unit, including: For the selected operation records, extract the maximum voltage value, minimum voltage value, and maximum voltage change rate of the individual battery during the charging and discharging process; Extract the initial temperature, maximum temperature, and time taken for the temperature to reach the peak value of the energy storage system during operation. The maximum voltage value, the minimum voltage value, the maximum voltage change rate, the starting temperature, the highest temperature, and the time taken for the temperature to reach its peak value are compared with the safety boundary values ​​in the energy storage unit's technical specifications. The operating conditions defined by the corresponding power grid state feature vector of those operating records where one or more parameters are close to or repeatedly touch the safety boundary value are marked as the characteristic operating condition mode. A feature label is established for each of the aforementioned characteristic operating conditions, and the feature label contains the range of major power grid characteristic parameters that cause stress.

10. The microgrid energy storage system for easy energy management connection according to claim 1, characterized in that, It also includes adaptive impedance identification of the grid connection point of the energy storage system, as detailed below: At the grid connection point of the energy storage system, a set of tiny current disturbance signals with a specific spectrum are injected; The voltage response signal at the grid connection point is measured synchronously during the injection of the small current disturbance signal; Based on the injected current disturbance signal and the measured voltage response signal, the equivalent impedance amplitude-frequency characteristic and phase-frequency characteristic viewed from the energy storage system side to the grid side are calculated. Based on the calculated equivalent impedance amplitude-frequency characteristics and phase-frequency characteristics, the equivalent short-circuit capacity and main resonant frequency points of the power grid are identified. The identified equivalent short-circuit capacity and main resonant frequency points are used as input parameters to assist in generating the power grid state feature vector or adjusting the strategy parameters in the energy storage control strategy library.

Images