New energy grid-connected optimization control method based on energy storage system

Abstract

The present application belongs to the technical field of light storage cooperative control, and discloses a new energy grid-connected optimization control method based on an energy storage system, which comprises the following steps: collecting power grid impedance fluctuation amplitude, harmonic distortion rate and environmental temperature data, obtaining an initial SOC value, correcting the SOC value, calculating the adjustable power margin and response delay parameter of the photovoltaic unit, and generating an efficiency responsibility table; constructing a response cluster, generating a frequency modulation task package, and then generating a three-dimensional efficiency mapping table; dynamically calculating inertia demand in combination with the power grid frequency change rate, automatically activating the responsibility transfer strategy in a high-altitude high-temperature scene, and generating an efficiency instruction set; analyzing response deviation by simulating typical disturbance of a high-altitude weak power grid scene, locating the source of deviation in combination with historical data, and dynamically adjusting the efficiency weight correction factor; updating the responsibility allocation rules of the efficiency responsibility table, dissolving low-efficiency clusters, and recombining high-adaptability units.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic-storage coordinated control technology, and more specifically, to a method for optimizing grid connection control of new energy sources based on energy storage systems. Background Technology

[0002] With the increasing penetration of renewable energy, photovoltaic-storage systems face challenges in high-altitude, weak power grids, such as delayed frequency regulation task allocation and insufficient conflict avoidance. Traditional methods rely on static rules and single parameters (such as SOC value), making it difficult to adapt to grid frequency fluctuations, sudden changes in harmonic distortion rate, and extreme environmental changes. Existing technologies mostly use preset models and fixed responsibility zones, lacking the integration of real-time environmental parameters (temperature, impedance fluctuations) and equipment status (lifespan degradation), resulting in insufficient system response efficiency and stability.

[0003] Existing solutions suffer from several core problems, including a disconnect between static models and dynamic scenarios, insufficient SOC estimation and lifetime compensation, incomplete conflict detection logic, limited dynamic correction capabilities, and poor resource coordination. Traditional methods fail to integrate real-time parameters, resulting in delayed frequency regulation task allocation; SOC estimation exhibits large errors in extreme temperature regions and lacks lifetime degradation correction, leading to frequency regulation deviations; conflict detection between photovoltaic reactive power commands and energy storage frequency regulation actions relies solely on symbol comparison, resulting in a high false negative rate; the system lacks a closed-loop optimization mechanism linking simulation and historical data, leading to delayed updates of efficiency weights; and the responsibility allocation strategy does not disband inefficient clusters based on efficiency scores, resulting in poor coordination between equipment response delays and lifetime compensation factors. Consequently, frequency regulation efficiency decreases, lifetime degradation in high-altitude weak grid scenarios, and stability risks increase. Summary of the Invention

[0004] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: a new energy grid-connected optimization control method based on an energy storage system, comprising:

[0005] S1: Deploy a sensor network to collect real-time data on grid impedance fluctuation amplitude, harmonic distortion rate, and ambient temperature, and obtain the initial SOC value of the energy storage unit. Correct the SOC value through the SOC estimation error correction mechanism, and then calculate the adjustable power margin and response delay parameters of the photovoltaic unit. Combine the lifetime decay coefficient and temperature compensation factor to generate an efficiency responsibility table that includes reactive power responsibility zoning, frequency regulation threshold, and lifetime compensation factor.

[0006] S2: Based on the performance responsibility table, construct a response cluster according to the adjustable power margin and response delay parameters, match the frequency modulation threshold with the response cluster type, generate frequency modulation task packages, allocate task execution rights through performance competition indicators, and generate a three-dimensional performance mapping table of time-frequency band-power.

[0007] S3: Based on the three-dimensional performance mapping table and performance responsibility table, and combined with the grid frequency change rate, dynamically calculate the inertia requirement, detect the conflict between photovoltaic reactive power instructions and energy storage frequency regulation actions, and perform cascade verification; automatically activate the responsibility transfer strategy in high-altitude and high-temperature scenarios to generate a performance instruction set that integrates reactive power compensation, frequency division power and dynamic inertia;

[0008] S4: Based on the performance instruction set, the response deviation is analyzed by simulating typical disturbances in high-altitude weak power grid scenarios, and the source of deviation is located by combining historical data, and the performance weight correction factor is dynamically adjusted.

[0009] S5: Based on the performance weight correction factor, update the responsibility allocation rules of the performance responsibility table, disband inefficient clusters, and reorganize highly adaptable units.

[0010] Furthermore, the method for correcting the SOC value includes:

[0011] By deploying a sensor network in the photovoltaic and energy storage area through a hybrid layout scheme, the grid impedance fluctuation amplitude, harmonic distortion rate, irradiance intensity and ambient temperature data of the photovoltaic units can be collected in real time.

[0012] The SOC value of the energy storage unit is obtained from the battery management system as the initial SOC value;

[0013] The collected data on power grid impedance fluctuation amplitude, harmonic distortion rate, ambient temperature, irradiance intensity, and initial SOC value were cleaned and aligned according to timestamps.

[0014] Set a normal temperature threshold. If the ambient temperature data is greater than or equal to the normal temperature threshold, use the initial SOC value directly as the SOC value.

[0015] If the ambient temperature data is lower than the normal temperature threshold, it is determined to be a low temperature environment, triggering the SOC estimation error correction mechanism to correct the initial SOC estimation value and obtain the corrected SOC value.

[0016] The SOC estimation error correction mechanism is as follows:

[0017] Define a reference temperature, and calculate the temperature compensation factor by combining ambient temperature data with the reference temperature.

[0018] The product of the temperature compensation factor and the initial SOC estimate is taken as the corrected SOC value.

[0019] Furthermore, the calculation method for the adjustable power margin and response delay parameters includes:

[0020] Set the normal range of SOC value and the default power suppression coefficient of adjustable power margin. Adjust the default power suppression coefficient according to the power regulation strategy based on the SOC value to obtain the adjusted power suppression coefficient.

[0021] The logic for adjusting the SOC value and power suppression coefficient is as follows:

[0022] If the SOC value is less than the minimum value of the normal range, the energy storage unit is determined to be in a low-charge state, and the power suppression coefficient is increased by a preset ratio; if the SOC value is greater than the maximum value of the normal range, the energy storage unit is determined to be in a high-charge state, and the power suppression coefficient is decreased by a preset ratio; if the SOC value is within the normal range, the default power suppression coefficient is maintained.

[0023] Based on the irradiance and ambient temperature, combined with the module efficiency and module area of ​​the photovoltaic unit, the theoretical maximum power of the photovoltaic unit under the current environment is calculated.

[0024] The grid harmonic limit for adjustable power margin is defined as setting a harmonic distortion threshold. If the harmonic distortion rate is less than or equal to the harmonic distortion threshold, the grid harmonic pollution is determined to be low, and the calculated theoretical maximum power remains unchanged.

[0025] If the harmonic distortion rate is greater than the harmonic distortion threshold, the power grid is judged to have high harmonic pollution, and the calculated theoretical maximum power is reduced by a preset ratio.

[0026] The instantaneous rate of change of photovoltaic unit power output is obtained as the power change rate. The range of adjustable power is dynamically adjusted according to the power change rate. Then, combined with the theoretical maximum power and power suppression coefficient obtained through grid harmonic limitation, the adjustable power margin is calculated.

[0027] Based on the amplitude of grid impedance fluctuation and ambient temperature, if the amplitude of grid impedance fluctuation is greater than the preset amplitude threshold, the grid is determined to be unstable. An impedance fluctuation correction factor is calculated using the amplitude of grid impedance fluctuation to correct the reference response time.

[0028] When the difference between the ambient temperature and the reference temperature is greater than the preset temperature deviation threshold, it is determined that the ambient temperature deviates from the reference temperature, and the temperature compensation factor is used to correct the reference response time.

[0029] The impedance fluctuation correction factor and temperature compensation factor are multiplied and superimposed on the reference response time to obtain the final response delay parameter.

[0030] Furthermore, the method for generating the performance accountability table includes:

[0031] Set the lifetime degradation coefficient of the energy storage unit. Based on the SOC value and ambient temperature data, correct the lifetime degradation coefficient according to the normal range of SOC and the normal temperature threshold. Take the product of the corrected lifetime degradation coefficient and the temperature compensation factor as the lifetime compensation factor.

[0032] Based on adjustable power margin, SOC value and response delay parameters, and according to the characteristics of high-altitude weak power grid environment, combined with power grid harmonic distortion rate and ambient temperature, the complete photovoltaic-storage area is divided into different types of sub-regions, and the reactive power responsibility of related equipment of photovoltaic unit and energy storage unit is dynamically allocated to obtain reactive power responsibility zoning.

[0033] Meanwhile, based on the SOC value and lifetime compensation factor, combined with adjustable power margin and response delay parameters, combined with threshold dynamic adjustment logic and different types of sub-region adaptation strategies, the frequency modulation threshold range of photovoltaic units and energy storage units is dynamically adjusted, thereby generating a dynamic frequency modulation threshold table.

[0034] The reactive power responsibility zoning, dynamic frequency regulation threshold table, and lifetime compensation factor are integrated to generate the final performance responsibility table.

[0035] Furthermore, the generation method of the frequency modulation task packet includes:

[0036] Based on the performance responsibility table, classification weights are set for adjustable power margin and response delay parameters using expert experience method, and then the classification score of the equipment is calculated using the classification score calculation formula.

[0037] Set a score threshold range and classify devices whose classification scores are greater than or equal to the maximum value of the score threshold range into high-frequency response clusters;

[0038] Devices whose classification scores are greater than or equal to the minimum value of the score threshold range but less than the score threshold range are classified as medium-frequency response clusters;

[0039] Devices with classification scores less than the minimum value of the score threshold range are classified into low-frequency response clusters;

[0040] Based on the divided response cluster types, according to the characteristics of different types of response clusters, the frequency modulation threshold and response cluster type are matched according to the task allocation logic to generate frequency modulation task packages for the corresponding frequency bands, including high-frequency frequency modulation tasks, medium-frequency frequency modulation tasks and low-frequency frequency modulation tasks.

[0041] The task allocation logic is defined as follows:

[0042] Real-time monitoring of power grid frequency change rate; setting a change rate threshold range; tasks with a frequency change rate greater than the maximum value of the change rate threshold range are designated as high-frequency frequency regulation tasks and are prioritized for execution by the high-frequency response cluster.

[0043] Tasks whose frequency change rate is greater than or equal to the minimum value of the change rate threshold interval and less than the maximum value of the change rate threshold interval are designated as medium-frequency sub-frequency modulation tasks and executed by calling the medium-frequency response cluster.

[0044] Tasks whose frequency change rate is less than the minimum value of the change rate threshold range are treated as low-frequency frequency modulation tasks and executed by calling the low-frequency response cluster.

[0045] Define the adjustable power margin range, and divide the adjustable power margin into three levels according to the margin range, each corresponding to a different type of response cluster.

[0046] Tasks that meet the corresponding frequency band but have a power requirement less than a certain percentage of the average power requirement of the corresponding frequency band are marked as hybrid task packages.

[0047] If a device's classification score matches the classification of the response cluster, but its corresponding adjustable power margin does not match the corresponding response cluster classification, then the device will be marked as a degraded device and will execute a hybrid task in the corresponding response cluster.

[0048] Furthermore, the method for generating the three-dimensional performance mapping table includes:

[0049] Performance competitiveness metrics include response delay parameters, adjustable power margin, and lifetime compensation factor.

[0050] The logic for allocating task execution rights is defined as follows:

[0051] By assigning weights to each performance competitiveness indicator using expert experience, and then weighting and integrating these indicators, the overall competitiveness score of the equipment is obtained.

[0052] Then, the comprehensive competition score is divided into three levels by threshold segmentation method, corresponding to high frequency frequency modulation tasks, medium frequency frequency modulation tasks and low frequency frequency modulation tasks respectively. Combined with the life compensation factor, the task execution rights of different frequencies are dynamically allocated to the equipment with the corresponding comprehensive competition score.

[0053] Set a fixed time window, record the timestamp of task triggering according to the time window as the time dimension, take the frequency band type of the task as the frequency band dimension, take the three levels of adjustable power margin as the power dimension, integrate the time dimension, frequency band dimension and power dimension to construct a three-dimensional performance mapping table.

[0054] Furthermore, the methods for calculating inertia requirements, detecting conflicts between photovoltaic reactive power commands and energy storage frequency regulation actions, and performing cascade verification include:

[0055] Based on the current frequency change rate of the power grid, the current frequency band task type is matched from the three-dimensional performance mapping table, and the response delay parameters of the corresponding equipment and the frequency modulation threshold of the current frequency band task are extracted. The required inertia is then calculated using the inertia requirement formula.

[0056] The current required inertia is compared with the preset minimum safe inertia. If the current required inertia is greater than or equal to the minimum safe inertia, it is marked as sufficient inertia. If the current required inertia is less than the minimum safe inertia, it is determined that the inertia is insufficient and the emergency frequency adjustment strategy is triggered.

[0057] The method for detecting conflicts between photovoltaic reactive power commands and energy storage frequency regulation actions is as follows:

[0058] Obtain the current photovoltaic reactive power command and energy storage demand frequency regulation action, and define the conflict condition between the photovoltaic reactive power command and energy storage demand frequency regulation action as follows:

[0059] The photovoltaic reactive power command requires an increase in reactive power, but the energy storage frequency regulation action requires a decrease in active power; the photovoltaic reactive power command requires a decrease in reactive power, but the energy storage frequency regulation action requires an increase in active power.

[0060] Compare the sign of the photovoltaic reactive power command with the direction of the energy storage frequency regulation action. If the sign and direction are opposite, it is determined to be a conflict; if the sign and direction are not opposite, it is determined to be a non-conflict.

[0061] The logic for performing cascaded verification includes inertia requirement verification, conflict resolution, and environmental adaptability verification.

[0062] Inertia requirement verification involves comparing the current required inertia with the rated inertia required for the current frequency band task. If the current required inertia is greater than or equal to a preset proportion of the rated inertia, it is determined that the requirement is met; if the current required inertia is less than a preset proportion of the rated inertia, it is determined that the requirement is not met.

[0063] The conflict handling mechanism is to prioritize adjusting the energy storage frequency regulation action (such as reducing power demand) to avoid conflicts if conflicts exist.

[0064] Environmental adaptability verification is performed in high-altitude weak power grid scenarios. If the lifetime compensation factor is less than the minimum value of the preset lifetime compensation threshold range, it is determined that the environment exceeds the limit, and the current frequency band task type is downgraded to the next lower frequency band task.

[0065] Furthermore, the generation method of the performance instruction set includes:

[0066] Based on the current frequency change rate and the current frequency band task type matched from the three-dimensional performance mapping table, responsibilities are allocated by judging responsibility priority logic;

[0067] The logic for prioritizing responsibility is defined as follows: for high-frequency frequency regulation tasks, frequency regulation is prioritized through the relevant equipment of the energy storage unit; for low-frequency frequency regulation tasks, reactive power compensation is prioritized through the relevant equipment of the photovoltaic unit.

[0068] For medium-frequency regulation tasks, the energy storage unit and photovoltaic unit related equipment are combined, with the energy storage unit related equipment regulating the active power and the photovoltaic unit related equipment synchronously regulating the reactive power.

[0069] Real-time monitoring of the performance of energy storage unit and photovoltaic unit related equipment; if the energy storage unit related equipment is not responding well, the active power regulation responsibility that the energy storage unit related equipment cannot handle is transferred to the photovoltaic unit related equipment through reactive power-active power coupling control.

[0070] Based on the responsibility allocation results, basic instructions are generated, including reactive power compensation instructions and frequency division power instructions;

[0071] The reactive power compensation command is calculated by combining the voltage change and lifetime compensation factor with the equipment's maximum reactive power capacity and voltage sensitivity coefficient to determine the reactive power adjustment amount, which is then used as the reactive power compensation command.

[0072] The responsibility allocation logic is as follows: if the responsibility is transferred to the photovoltaic unit, the reactive power increment is generated through reactive power compensation instructions based on the voltage change calculated by the AVC system and directly sent to the relevant equipment of the photovoltaic unit; if the responsibility is transferred to the energy storage unit, the priority of the photovoltaic reactive power instructions is reduced, and the voltage is compensated through the energy storage system first.

[0073] The frequency division power command is based on the frequency change rate and lifetime compensation factor, combined with the equipment's maximum power calculation and the active power adjustment amount of the frequency response sensitivity coefficient.

[0074] The responsibility allocation logic is as follows: if the responsibility is transferred to the energy storage unit, the energy storage frequency regulation power is calculated based on the frequency change rate and response delay parameters; if the responsibility is transferred to the photovoltaic unit, the priority of the energy storage frequency regulation command is reduced, and the active power is adjusted through the relevant equipment of the photovoltaic unit first.

[0075] The basic instructions are compensated and modified based on environmental conditions and lifespan. The modification logic is as follows:

[0076] For high-altitude and high-temperature scenarios, if high altitude is detected, the energy storage frequency regulation power will be reduced proportionally based on the lifespan compensation factor.

[0077] Based on the device's classification score, if the device is classified as a low-frequency response cluster, the current frequency band task type will be forcibly downgraded, and the frequency modulation threshold will be adjusted proportionally.

[0078] In high-altitude and high-temperature scenarios, the dynamic inertia constraint is further integrated, and the corrected logic is as follows:

[0079] If the current required inertia is greater than the minimum safe inertia, reduce the energy storage frequency regulation power proportionally.

[0080] If the reactive power compensation command and the frequency division power command conflict in direction, the reactive power compensation command shall be given priority and the frequency division power weight shall be dynamically adjusted.

[0081] The constraints for defining the performance instruction set are reactive power compensation priority, frequency modulation power constraint, and dynamic inertia guarantee.

[0082] The final reactive power compensation command and frequency division power command are integrated to generate an efficiency command set that meets the constraints.

[0083] Furthermore, the method for dynamically adjusting the performance weight correction factor includes:

[0084] Based on the performance instruction set, a simulation model is constructed according to high-altitude environmental parameters, weak power grid characteristics and typical disturbance types to simulate different types of typical disturbances;

[0085] In the simulation environment, reactive power compensation commands and frequency division power commands are executed, the actual response is recorded, and then the response deviation is calculated.

[0086] By combining response deviations with historical data, regression analysis is used to pinpoint the main causes of the deviations, and the performance weighting factors, including the frequency response sensitivity coefficient and the voltage sensitivity coefficient, are dynamically adjusted.

[0087] Furthermore, the methods for updating the performance responsibility table, disbanding inefficient clusters, and reorganizing highly adaptable units include:

[0088] Based on the revised performance weighting factor, the performance score of each device is calculated using the performance scoring formula.

[0089] The responsibility allocation rules are updated based on the equipment's performance rating.

[0090] Set a performance score threshold, mark devices with performance scores below the threshold as inefficient devices, mark clusters with a certain proportion of inefficient devices as inefficient clusters, disband inefficient clusters and remove responsibility assignments.

[0091] Devices with performance scores greater than or equal to the performance score threshold are marked as high-efficiency devices, and a list of high-efficiency devices is generated.

[0092] Based on a list of efficient devices, clusters are reorganized according to geographical proximity and functional complementarity of the devices, and the cluster combination is optimized through dynamic clustering algorithms to ensure responsiveness.

[0093] The technical effects and advantages of the new energy grid-connected optimization control method based on energy storage system of this invention are as follows:

[0094] This invention calculates a temperature compensation factor by combining the difference between ambient temperature and reference temperature, thereby achieving dynamic correction of SOC under low / high temperature environments and significantly improving estimation accuracy. At the same time, it introduces a lifetime decay coefficient to generate a lifetime compensation factor, dynamically correcting the SOC and adjustable power margin of the energy storage unit, extending equipment life and enhancing frequency regulation response reliability.

[0095] Secondly, based on adjustable power margin, response delay parameters, and lifetime compensation factors, high-frequency / medium-frequency / low-frequency response clusters are autonomously divided using expert experience methods to achieve refined matching of task packages (frequency tuning requirements); further, time, frequency band, and power dimensions are integrated to construct a three-dimensional performance mapping table to improve the real-time performance and coordination of task execution allocation.

[0096] Then, in high-altitude weak grid scenarios, the task type is automatically downgraded and the frequency regulation threshold is adjusted through a responsibility transfer strategy to avoid equipment overload; at the same time, the conflict between photovoltaic reactive power commands and energy storage frequency regulation actions is determined by comparing the signs and directions, and the energy storage actions are adjusted first to avoid risks. The system stability is ensured by combining inertia demand verification and environmental adaptability verification.

[0097] Next, by analyzing response deviations (such as SOC estimation error and harmonic distortion rate) through simulation of typical disturbances in high-altitude weak power grids and linking them with historical data, the efficiency weight correction factors (such as frequency response sensitivity coefficient and voltage sensitivity coefficient) are dynamically adjusted to form a closed-loop optimization mechanism.

[0098] Finally, based on the revised performance scoring formula, inefficient clusters are disbanded and highly adaptable units are reorganized. Dynamic clustering optimization is driven by quantitative indicators, which significantly improves the overall system synergy and resource utilization.

[0099] This invention solves the problems of large SOC estimation error, rigid task allocation, and insufficient conflict handling in traditional photovoltaic-storage systems in high-altitude weak power grids through a full-process design of "dynamic correction → responsibility allocation → conflict avoidance → closed-loop optimization → performance reorganization", and achieves a systematic improvement in frequency regulation efficiency, equipment life and power grid stability. Attached Figure Description

[0100] Figure 1 This is a schematic diagram of the new energy grid-connected optimization control method based on energy storage system of the present invention;

[0101] Figure 2 This is a schematic diagram of the new energy grid-connected optimization control method based on energy storage system of the present invention;

[0102] Figure 3 This is a schematic diagram of the new energy grid-connected optimization control system based on the energy storage system of the present invention. Detailed Implementation

[0103] 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 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.

[0104] Example 1

[0105] Please see Figure 1 and Figure 2 As shown in this embodiment, the new energy grid-connected optimization control method based on energy storage system includes:

[0106] S1: Deploy a sensor network to collect real-time data on grid impedance fluctuation amplitude, harmonic distortion rate, and ambient temperature, and obtain the initial SOC value of the energy storage unit. Correct the SOC value through the SOC estimation error correction mechanism, and then calculate the adjustable power margin and response delay parameters of the photovoltaic unit. Combine the lifetime decay coefficient and temperature compensation factor to generate an efficiency responsibility table that includes reactive power responsibility zoning, frequency regulation threshold, and lifetime compensation factor.

[0107] S2: Based on the performance responsibility table, construct a response cluster according to the adjustable power margin and response delay parameters, match the frequency modulation threshold with the response cluster type, generate frequency modulation task packages, allocate task execution rights through performance competition indicators, and generate a three-dimensional performance mapping table of time-frequency band-power.

[0108] S3: Based on the three-dimensional performance mapping table and performance responsibility table, and combined with the grid frequency change rate, dynamically calculate the inertia requirement, detect the conflict between photovoltaic reactive power instructions and energy storage frequency regulation actions, and perform cascade verification; automatically activate the responsibility transfer strategy in high-altitude and high-temperature scenarios to generate a performance instruction set that integrates reactive power compensation, frequency division power and dynamic inertia;

[0109] S4: Based on the performance instruction set, the response deviation is analyzed by simulating typical disturbances in high-altitude weak power grid scenarios, and the source of deviation is located by combining historical data, and the performance weight correction factor is dynamically adjusted.

[0110] S5: Based on the performance weight correction factor, update the responsibility allocation rules of the performance responsibility table, disband inefficient clusters, and reorganize highly adaptable units;

[0111] By deploying a sensor network in the photovoltaic-storage hybrid area through a hybrid layout scheme, the system can collect real-time data on the grid impedance fluctuation amplitude of the photovoltaic units (using high-precision sensors such as broadband measurement terminals to collect impedance data in real time), harmonic distortion rate (using harmonic analyzers or multi-functional sensors to monitor harmonic distortion rate (THD value) in real time), irradiance (which can be obtained from weather stations via API or predicted from historical data), and ambient temperature data (which can be collected by temperature sensors or obtained by reading weather station data via API).

[0112] The hybrid layout scheme involves evenly distributing sensor nodes across the target area for large-area photovoltaic power plants, ensuring that each area has a monitoring capability. For areas with higher monitoring requirements, such as those near grid access points or high-risk areas, the sensor density is increased to improve monitoring accuracy and response speed. This approach can improve the monitoring accuracy of key areas while ensuring coverage.

[0113] The initial SOC value is obtained from the SOC value of the energy storage unit (the state of charge, which reflects the remaining capacity of the energy storage unit) obtained from the battery management system (BMS).

[0114] The collected data on power grid impedance fluctuation amplitude, harmonic distortion rate, ambient temperature, irradiance, and initial SOC value were cleaned (abnormal data was removed, and missing data was filled in) and aligned by timestamp.

[0115] Set a normal temperature threshold. If the ambient temperature data is greater than or equal to the normal temperature threshold, use the initial SOC value directly as the SOC value.

[0116] If the ambient temperature data is lower than the normal temperature threshold, it is determined to be a low temperature environment, triggering the SOC estimation error correction mechanism to correct the initial SOC estimation value and obtain the corrected SOC value.

[0117] The SOC estimation error correction mechanism is as follows:

[0118] Define a reference temperature, and calculate the temperature compensation factor by combining ambient temperature data with the reference temperature.

[0119] The method for calculating the temperature compensation factor is as follows:

[0120] ;

[0121] The temperature correction factor is an empirical factor for correcting battery performance based on temperature, typically 0.05, and the reference temperature is a predefined optimal ambient temperature (e.g., 25 degrees Celsius as the reference temperature).

[0122] The product of the temperature compensation factor and the initial SOC estimate is taken as the corrected SOC value;

[0123] If the SOC value is greater than 100%, it is truncated to 100%; if the SOC value is less than 0%, it is truncated to 0%. That is, the minimum SOC value is defined as 0%, and the maximum value is defined as 100%.

[0124] Set the normal range of SOC value (e.g., [20%, 80%]) and the default power suppression coefficient of adjustable power margin (e.g., 1). Adjust the default power suppression coefficient according to the power regulation strategy based on the SOC value to obtain the adjusted power suppression coefficient.

[0125] The logic for adjusting the SOC value and power suppression coefficient is as follows:

[0126] If the SOC value is less than the minimum value of the normal range, the energy storage unit is determined to be in a low-charge state, and the power suppression coefficient is increased by a preset ratio (e.g., 20%) (e.g., 1.2) to limit the photovoltaic output power and avoid over-discharge; if the SOC value is greater than the maximum value of the normal range, the energy storage unit is determined to be in a high-charge state, and the power suppression coefficient is decreased by a preset ratio (e.g., 0.8) to limit the photovoltaic output power and avoid the risk of overcharging; if the SOC value is within the normal range, the default power suppression coefficient is maintained.

[0127] Based on the irradiance and ambient temperature, combined with the module efficiency and module area of ​​the photovoltaic unit, the theoretical maximum power of the photovoltaic unit under the current environment is calculated.

[0128] Theoretical maximum power of photovoltaic unit The calculation formula is:

[0129] ;

[0130] in, Indicates irradiation intensity. Indicates the component area. Indicates component efficiency. This represents the temperature coefficient (typically 0.005 / ℃). This represents ambient temperature data. Indicates the reference temperature;

[0131] The grid harmonic limit for adjustable power margin is defined as setting a harmonic distortion threshold. If the harmonic distortion rate is less than or equal to the harmonic distortion threshold, the grid harmonic pollution is determined to be low, and the calculated theoretical maximum power remains unchanged.

[0132] If the harmonic distortion rate is greater than the harmonic distortion threshold, the power grid is considered to have high harmonic pollution. The calculated theoretical maximum power is reduced by a preset ratio (e.g., 20%) to reduce interference to the power grid. For example, the theoretical maximum power is limited to 80% of its original value.

[0133] The instantaneous rate of change of photovoltaic unit power output is obtained as the power change rate. The range of adjustable power is dynamically adjusted according to the power change rate. Then, combined with the theoretical maximum power and power suppression coefficient obtained through grid harmonic limitation, the adjustable power margin is calculated.

[0134] Adjustable power margin The calculation formula is: ;

[0135] in, Indicates the power suppression coefficient. Indicates the sampling time interval (e.g., 1 second). It represents the rate of change of power, reflecting the severity of power fluctuations. The calculation method is to calculate the slope of power change through historical power data (such as the sampled value of the past minute). That is, take the power value of the next sampling time interval and subtract the previous power value, and calculate the ratio of the difference result to the sampling time interval. Used to assess the risk of power surges and adjust the theoretical maximum power using a power suppression coefficient, thereby avoiding equipment damage or grid disturbances caused by surges; in the formula, This is a baseline value, representing the theoretical maximum regulation capability of a photovoltaic unit. It is a dynamic adjustment factor used to suppress the actual adjustable power based on the severity of power fluctuations;

[0136] The calculation logic for adjustable power margin is that the larger the power change rate, the more drastic the photovoltaic output fluctuation, which needs to be determined by the power suppression coefficient and... The product of the two values ​​reduces the adjustable power margin, thereby avoiding grid disturbances or equipment damage caused by sudden changes. If the calculated result of the adjustable power margin is negative, it indicates that the current power fluctuation is too large, and the suppression strategy needs to be further adjusted (such as extending the sampling time or reducing the power suppression coefficient).

[0137] Based on the amplitude of grid impedance fluctuation and ambient temperature, if the amplitude of grid impedance fluctuation is greater than the preset amplitude threshold (e.g., 5Ω), the grid is determined to be unstable. An impedance fluctuation correction factor is calculated based on the amplitude of grid impedance fluctuation to correct the reference response time.

[0138] For example, a formula for calculating an impedance fluctuation correction factor is as follows:

[0139] ;in, This represents the impedance fluctuation correction factor. Indicates the amplitude of grid impedance fluctuation. Indicates the amplitude threshold (e.g., 5Ω);

[0140] It should be noted that when the amplitude of the power grid impedance fluctuation is greater than the preset amplitude threshold, the response delay will increase linearly with the impedance fluctuation (for example, the greater the impedance fluctuation, the higher the delay).

[0141] When the difference between the ambient temperature and the reference temperature is greater than the preset temperature deviation threshold (such as 5℃ or 10℃), it is determined that the ambient temperature deviates from the reference temperature, and the temperature compensation factor is used to correct the reference response time.

[0142] It should be noted that when the ambient temperature deviates from the reference temperature (e.g., 25°C), the response delay will be dynamically adjusted according to the temperature change.

[0143] Low temperature environments (e.g., below 25°C): Battery performance degrades, and response delay is appropriately prolonged;

[0144] High-temperature environments (e.g., above 25°C): Battery performance is improved, and response latency is slightly reduced;

[0145] The impedance fluctuation correction factor and temperature compensation factor are multiplied and superimposed on the reference response time to obtain the final response delay parameter (used to represent the time delay required to complete a frequency regulation action (such as the time from receiving a command to the actual power output of an energy storage system)).

[0146] Specifically, this involves multiplying the reference response time, impedance fluctuation correction factor, and temperature correction factor together.

[0147] Because the effects of impedance fluctuations and temperature on response delay are independent and superimposed, the cumulative effect of both must be considered simultaneously.

[0148] For example, if the grid impedance fluctuates greatly (requiring a longer delay) and the ambient temperature is low (also requiring a longer delay), the combined effect of both will significantly increase the final delay.

[0149] Impedance fluctuation impact: The more unstable the power grid (the larger the impedance fluctuation), the higher the response delay. Avoid exacerbating power grid disturbances due to frequent adjustments.

[0150] Temperature effect: When the ambient temperature deviates from the standard value, the response delay is dynamically adjusted according to the temperature coefficient to ensure the stability of the system under extreme temperatures;

[0151] Set up a real-time update mechanism: The parameters of all correction factors (such as grid impedance and ambient temperature) need to be re-evaluated periodically to ensure that the response delay parameters can adapt to the real-time changes in the environment and grid conditions;

[0152] Set the lifetime degradation coefficient of the energy storage unit. Based on the SOC value and ambient temperature data, correct the lifetime degradation coefficient according to the normal range of SOC and the normal temperature threshold. Take the product of the corrected lifetime degradation coefficient and the temperature compensation factor as the lifetime compensation factor.

[0153] The logic for defining and correcting the lifetime degradation coefficient is as follows: First, set a default lifetime degradation coefficient, such as 1. If the SOC value is not within the normal range, the lifetime degradation coefficient needs to be significantly reduced. The default lifetime degradation coefficient can be reduced proportionally, such as by 10%. In this case, the corrected lifetime degradation coefficient is 0.9.

[0154] If the ambient temperature data is greater than the normal temperature threshold, it is judged as a high-temperature environment, and the lifespan degradation coefficient is reduced proportionally, such as by 10%.

[0155] If the SOC value is not within the normal range and is in a high-temperature environment, first reduce the life decay coefficient proportionally, and then further reduce the reduced life decay coefficient proportionally.

[0156] The physical meaning of the lifespan compensation factor is to reflect the overall lifespan health status under the current SOC value and ambient temperature data;

[0157] Based on adjustable power margin, SOC value and response delay parameters, and according to the characteristics of high-altitude weak power grid environment, combined with power grid harmonic distortion rate and ambient temperature, the complete photovoltaic-storage area is divided into different types of sub-regions, and the reactive power responsibility of related equipment of photovoltaic unit and energy storage unit is dynamically allocated to obtain reactive power responsibility zoning.

[0158] Specifically, the characteristics of high-altitude weak power grid environments include poor power grid stability, low temperature, and high harmonic distortion (THD).

[0159] The photovoltaic-energy storage hybrid area is divided into three sub-regions (high THD area, low THD area, and grid fluctuation sensitive area).

[0160] In high THD areas (such as industrial load areas), SVG or inverters (photovoltaic units) should be prioritized because they have dynamic response capability (5-20ms) and harmonic filtering function, which can quickly suppress harmonics and provide reactive power support.

[0161] In low THD areas (such as commercial load areas), parallel capacitors or energy storage PCS (energy storage units) are used to meet basic reactive power requirements through static compensation or dynamic adjustment.

[0162] In areas sensitive to grid fluctuations (such as high-altitude weak grid nodes), energy storage PCS assumes the main responsibility for reactive power support because its millisecond-level response capability can quickly balance grid fluctuations.

[0163] The reactive power responsibility allocation logic for the equipment is as follows:

[0164] For photovoltaic units and energy storage units, the available reactive capacity of the equipment is calculated based on the adjustable power margin, SOC value and temperature compensation factor. If the available reactive capacity is lower than a certain percentage (such as 70%) of the rated capacity, the reactive responsibility allocation ratio is restricted.

[0165] The response delay parameter is used to screen whether the equipment can handle high-frequency, fast-response reactive power regulation tasks;

[0166] Meanwhile, based on the SOC value and lifetime compensation factor, combined with adjustable power margin and response delay parameters, combined with threshold dynamic adjustment logic and different types of sub-region adaptation strategies, the frequency modulation threshold range of photovoltaic units and energy storage units is dynamically adjusted, thereby generating a dynamic frequency modulation threshold table (including sub-region ID, upper frequency limit, lower frequency limit, frequency modulation threshold, adjustable power margin, SOC value and delay response parameters).

[0167] Specifically, the threshold dynamic adjustment logic is to set a default frequency regulation threshold range (such as ±0.15Hz) based on the power grid stability requirements.

[0168] If the SOC value of the energy storage unit is not within the normal range (e.g., less than 20% or greater than 80%), calculate the product of the SOC value and the lifetime compensation factor to obtain the current available power (e.g., lifetime compensation factor = 0.85 at high temperature, available power = 70%), and proportionally relax the frequency regulation threshold (e.g., relax to ±0.2Hz) to limit its frequent charging and discharging.

[0169] If the SOC value is within the normal range (e.g., greater than or equal to 20% and less than or equal to 80%) and the ambient temperature does not deviate from the reference temperature, the default frequency modulation threshold range will be maintained to ensure rapid response capability.

[0170] Set a response delay threshold. For devices with response delay parameters greater than the response delay threshold (such as traditional inverters with response time > 50ms), the frequency modulation threshold is relaxed proportionally (e.g., relaxed to ±0.2Hz).

[0171] For devices with response delay parameters less than or equal to the response delay threshold (such as energy storage PCS with a response time <5ms), the threshold can be tightened proportionally (e.g., tightened to ±0.1Hz) to improve frequency modulation accuracy.

[0172] Set the adjustable power margin range (e.g., [70%, 90%]). If the adjustable power margin is greater than or equal to the maximum value of the margin range (the device has sufficient available power), maintain the default frequency modulation threshold range.

[0173] If the adjustable power margin is greater than or equal to the minimum value of the margin range and less than the maximum value of the margin range (the available power of the equipment is limited), the frequency modulation threshold is relaxed proportionally (e.g., relaxed to ±0.2Hz).

[0174] If the adjustable power margin is less than the minimum value of the margin range (the equipment is nearing its life limit), the frequency modulation threshold is further relaxed proportionally (e.g., relaxed to ±0.25Hz), and the charging and discharging frequency is limited;

[0175] The sub-region threshold strategy is as follows:

[0176] For high THD areas (such as industrial load areas): if the adjustable power margin is greater than or equal to the maximum value of the margin range, the frequency regulation threshold is tightened proportionally (e.g., tightened to ±0.1Hz), and the energy storage PCS takes the lead in undertaking high-frequency regulation tasks.

[0177] If the adjustable power margin is less than the minimum value of the margin range, the frequency modulation threshold will remain unchanged within the default frequency modulation threshold range.

[0178] For low THD areas (such as commercial load areas): the frequency regulation threshold is relaxed proportionally or the default frequency regulation threshold range is maintained unchanged (such as ±0.15Hz), and the photovoltaic inverter and parallel capacitor work together to complete the basic frequency regulation;

[0179] For areas sensitive to power grid fluctuations (such as high-altitude weak power grid nodes): obtain the real-time frequency change rate of the power grid, and dynamically adjust the threshold according to the real-time frequency change rate of the power grid (e.g., when the real-time frequency change rate of the power grid is >0.5Hz / s, trigger emergency frequency regulation, and automatically tighten the threshold to the minimum by the maximum proportion (e.g., tighten to ±0.05Hz)).

[0180] Exemplary scenario description:

[0181] Scenario 1: A certain energy storage unit has a SOC of 90% (abnormal range), an ambient temperature of 40℃ (high temperature), a lifespan degradation coefficient of 0.85, a usable power of 70% of the rated power (adjustable power margin of 70%), and a frequency modulation threshold relaxed to ±0.2Hz;

[0182] Scenario 2: A photovoltaic inverter has a SOC of 50% (normal range), an ambient temperature of -10℃ (low temperature), and after SOC correction, the usable power is 95% of the rated power (adjustable power margin = 95%). The frequency modulation threshold is maintained at ±0.15Hz, and the frequency deviation is dynamically compensated by SVG.

[0183] The reactive power responsibility zoning, dynamic frequency regulation threshold table, and lifetime compensation factor are integrated to generate the final performance responsibility table.

[0184] Establish a dynamic update mechanism: All parameters of the performance accountability table must be monitored in real time and updated regularly to ensure the adaptability of the performance accountability table;

[0185] It should be noted that, through the above process, the performance responsibility table can comprehensively reflect the equipment status (SOC), regulation capability (adjustable power margin), and response characteristics (delay parameters), thereby realizing a dynamically optimized power grid regulation strategy.

[0186] Based on the performance responsibility table, classification weights are set for adjustable power margin and response delay parameters using expert experience (e.g., adjustable power margin is 0.6, response delay parameter is 0.4), and then the classification score of the equipment is calculated using the classification score calculation formula.

[0187] The formula for calculating the classification score is: ;

[0188] in, Indicates the category score. The classification weights represent the adjustable power margin. Indicates adjustable power margin. The classification weights represent the response delay parameter. This represents the response delay parameter;

[0189] Set a score threshold range (e.g., 70%-90%), and classify devices whose classification scores are greater than or equal to the maximum value of the score threshold range into high-frequency response clusters;

[0190] Devices whose classification scores are greater than or equal to the minimum value of the score threshold range but less than the score threshold range are classified as medium-frequency response clusters;

[0191] Devices with classification scores less than the minimum value of the score threshold range are classified into low-frequency response clusters;

[0192] Based on the divided response cluster types, according to the characteristics of different types of response clusters, the frequency modulation threshold and response cluster type are matched according to the task allocation logic to generate frequency modulation task packages for the corresponding frequency bands, including high-frequency frequency modulation tasks, medium-frequency frequency modulation tasks and low-frequency frequency modulation tasks.

[0193] The task allocation logic is defined as follows:

[0194] Real-time monitoring of power grid frequency change rate; setting a change rate threshold range; tasks with a frequency change rate greater than the maximum value of the change rate threshold range are designated as high-frequency frequency regulation tasks and are prioritized for execution by the high-frequency response cluster.

[0195] Tasks whose frequency change rate is greater than or equal to the minimum value of the change rate threshold interval and less than the maximum value of the change rate threshold interval are designated as medium-frequency sub-frequency modulation tasks and executed by calling the medium-frequency response cluster.

[0196] Tasks whose frequency change rate is less than the minimum value of the change rate threshold range are treated as low-frequency frequency modulation tasks and executed by calling the low-frequency response cluster.

[0197] Define the adjustable power margin range, and divide the adjustable power margin into three levels according to the margin range, each corresponding to a different type of response cluster.

[0198] Specifically, adjustable power margins greater than or equal to the maximum value of the margin range are assigned to high-frequency response clusters, adjustable power margins greater than or equal to the minimum value of the margin range and less than the maximum value of the margin range are assigned to mid-frequency response clusters, and adjustable power margins less than the minimum value of the margin range are assigned to low-frequency response clusters.

[0199] Tasks that meet the corresponding frequency band but have a power requirement less than a certain percentage of the average power requirement of the corresponding frequency band are marked as hybrid task packages.

[0200] For example, if the average power requirement of a medium-frequency frequency modulation task is 100kW and the minimum ratio is set to 40%, and the power requirement of a certain medium-frequency frequency modulation task is 60kW, which is 40% less than 100kW, then it is marked as a medium-frequency hybrid task.

[0201] If the device's classification score matches the classification of the response cluster, but the corresponding adjustable power margin does not match the corresponding response cluster classification, then the device will be marked as a degraded device and will execute a hybrid task in the corresponding response cluster.

[0202] Through a dynamic adjustment mechanism, the response cluster partitioning and task package allocation strategies are automatically updated based on the real-time status of the equipment (such as a decrease in adjustable power margin), ensuring the reliability of frequency modulation task execution and the safety of equipment operation.

[0203] Example Scenario

[0204] Scenario 1: Adjustable power margin = 65%, response delay parameter = 20ms → Intermediate frequency cluster (degradation mode) → Hybrid intermediate frequency task package (±0.15Hz, power requirement 60kW) → Power requirement reduced to 60% of rated capacity, frequency modulation threshold relaxed to ±0.18Hz;

[0205] Scenario 2: Adjustable power margin = 85%, response latency = 10ms → High-frequency cluster → High-frequency task package (±0.1Hz, power requirement 100kW) → Prioritize execution of high-frequency frequency modulation tasks;

[0206] Scenario 3: Adjustable power margin = 60%, response latency = 60ms → Low-frequency cluster → Low-frequency frequency modulation task (±0.2Hz, power requirement 50kW) → Only execute basic frequency modulation task;

[0207] The performance competition metrics are defined as follows: response latency parameters (quantifying task allocation weights based on equipment response latency parameters (e.g., <5ms for high priority, >50ms for low priority)), adjustable power margin (assessing equipment availability based on adjustable power margin (e.g., ≥90% for high priority, <70% for low priority)), and lifetime compensation factor (using the lifetime compensation factor as the available power attenuation coefficient (e.g., 0.85 = high loss, 0.95 = low loss), directly affecting the task allocation strategy).

[0208] The logic for allocating task execution rights is defined as follows:

[0209] By using expert experience to assign weights to each performance competitiveness indicator (e.g., response speed accounts for 40%, power margin accounts for 40%, and lifespan compensation factor accounts for 20%), the weighted and integrated performance competitiveness indicators are then combined to obtain the overall competitiveness score of the equipment.

[0210] The weighted fusion method is as follows: Overall score = weight 1 × response speed weight + weight 2 × power margin weight + weight 3 × lifetime compensation factor weight; where weight 1, weight 2 and weight 3 are the weights set for each performance competition indicator.

[0211] Then, the comprehensive competition score is divided into three levels by threshold segmentation method, corresponding to high frequency frequency modulation tasks, medium frequency frequency modulation tasks and low frequency frequency modulation tasks respectively. Combined with the life compensation factor, the task execution rights of different frequencies are dynamically allocated to the equipment with the corresponding comprehensive competition score.

[0212] Specifically, a comprehensive competition score threshold range (e.g., 0.7-0.9) and a life compensation factor threshold range (e.g., 0.8-0.9) are set.

[0213] Devices with a comprehensive competitive score greater than or equal to the maximum value of the comprehensive score threshold range (e.g., 0.9) are assigned to high-frequency frequency modulation tasks for priority execution. If the device's lifespan compensation factor is less than the maximum value of the lifespan compensation threshold range, the margin range of the adjustable power margin is dynamically widened (e.g., allowing 70%-85% of devices to be degraded for execution).

[0214] Devices with a comprehensive score greater than or equal to the minimum value of the comprehensive score threshold range and less than the maximum value of the comprehensive score threshold range are assigned to the medium-frequency frequency modulation task for priority execution. If the lifespan compensation factor is less than or equal to the maximum value of the lifespan compensation threshold range and greater than or equal to the minimum value of the lifespan compensation threshold range, then the device is switched to the low-frequency frequency modulation task.

[0215] Devices with a comprehensive score less than the minimum value of the comprehensive score threshold range are assigned to low-frequency frequency modulation tasks. If the lifespan compensation factor is less than the minimum value of the lifespan compensation threshold range, the task is paused and the preset backup device is activated.

[0216] Set a fixed time window and record the timestamp of task triggering according to the time window as the time dimension (example: [timestamp; task type; assigned device; status]; [2000-01-30 12:00; high-frequency task; device A; success]). Use the frequency band type of the task as the frequency band dimension (example: [frequency band type; frequency change rate range; corresponding frequency modulation threshold]; [high-frequency task; >0.5Hz / s; ±0.1Hz]). Use the three levels of adjustable power margin as the power dimension (example: [power margin; frequency modulation accuracy; device applicability]; [≥90%; ±0.1Hz; high-frequency cluster]). Integrate the time dimension, frequency band dimension, and power dimension to construct a three-dimensional performance mapping table.

[0217] The mapping table is monitored and updated periodically. If the power grid frequency change rate changes abruptly (e.g., >0.5Hz / s), a high-frequency task package is immediately triggered and the power demand is adjusted.

[0218] The task package parameters are automatically adjusted based on the real-time status of the equipment (such as a decrease in adjustable power margin or an increase in lifespan compensation factor).

[0219] Example of a three-dimensional performance accountability table:

[0220] [Timestamp; Frequency band type; Adjustable power margin; Frequency modulation threshold; Response delay parameter; Device ID; Status;]

[0221] [2000-01-30 12:00; High-frequency tuning task; 95%; ±0.1Hz; <5ms; A001; Success]

[0222] [2000-01-30 12:05; Medium frequency modulation task; 85%; ±0.15Hz; <15ms; B002; Success]

[0223] [2000-01-30 12:10; Low-frequency tuning task; 60%; ±0.2Hz; <30ms; C003; Pause (insufficient power)];

[0224] In high-altitude and high-temperature scenarios, by automatically activating the responsibility transfer strategy, and combining reactive power compensation, frequency division power demand and dynamic inertia constraints, an efficiency instruction set (ΔQ, ΔP) that integrates multiple dimensions of objectives is generated to ensure equipment lifespan safety and grid stability.

[0225] Based on the current frequency change rate of the power grid, the current frequency band task type (e.g., Δf / dt=0.6Hz / s → high frequency modulation task) is matched from the three-dimensional performance mapping table, and the response delay parameters of the corresponding equipment and the frequency modulation threshold of the current frequency band task (used to represent the maximum frequency deviation range allowed for the current frequency band task) are extracted. The required inertia is then calculated using the inertia requirement formula.

[0226] The formula for calculating inertia requirement is: ;in, This indicates the current required inertia. Indicates the power requirement of the current task (e.g.) =100kW), This indicates the frequency modulation threshold for the current frequency band task;

[0227] The current required inertia is compared with the preset minimum safe inertia. If the current required inertia is greater than or equal to the minimum safe inertia, it is marked as sufficient inertia. If the current required inertia is less than the minimum safe inertia, it is determined that the inertia is insufficient, and an emergency frequency regulation strategy is triggered (such as starting the backup energy storage unit).

[0228] The calculation of the minimum safe inertia is obtained by dynamic optimization based on the rate of change of frequency. Specifically:

[0229] Obtain key parameters of system operation and quantify the current inertia level of the system, including synchronous machine inertia, photovoltaic virtual inertia, and total system inertia; calculate the maximum frequency change rate using the frequency change rate formula; if the maximum frequency change rate is greater than the protection threshold of the equipment, the system inertia needs to be increased (e.g., through energy storage or virtual inertia support), and then the total system inertia is recalculated until the maximum frequency change rate is less than or equal to the protection threshold. The adjusted total system inertia is then used as the minimum safe inertia.

[0230] The method for detecting conflicts between photovoltaic reactive power commands and energy storage frequency regulation actions is as follows:

[0231] Obtain the current photovoltaic reactive power command (the photovoltaic reactive power command is the real-time reactive power adjustment target value issued by the AVC system (Automatic Voltage Control) to maintain grid voltage stability) and energy storage demand frequency regulation action (energy storage frequency regulation action is the active power adjustment command triggered by the energy storage coordination controller (such as the AGC system) based on the grid frequency deviation or frequency change rate), and define the conflict condition between the photovoltaic reactive power command and the energy storage demand frequency regulation action as follows:

[0232] The photovoltaic reactive power command requires an increase in reactive power (e.g., +1.2MVar), but the energy storage frequency regulation action requires a decrease in active power (e.g., a reduction in SOC); the photovoltaic reactive power command requires a decrease in reactive power (e.g., -0.8MVar), but the energy storage frequency regulation action requires an increase in active power (e.g., an increase in SOC).

[0233] Compare the sign (+ / -) of the photovoltaic reactive power command with the direction (+ / -) of the energy storage frequency regulation action (e.g., ΔP_sto=-20kW indicates a reduction in active power). If the sign and direction are opposite, it is determined to be a conflict; if the sign and direction are not opposite, it is determined to be a non-conflict.

[0234] Example:

[0235] Photovoltaic reactive power command: +1.2MVar (increase reactive power);

[0236] Energy storage frequency regulation action: ΔP_sto = -20kW (reduce active power);

[0237] Decision: Conflict (opposite signs);

[0238] It should be noted that by integrating the current task type with the real-time collected photovoltaic reactive power commands and energy storage frequency regulation action data, it is possible to determine whether precise coordinated control can be achieved.

[0239] The logic for performing cascaded verification includes inertia requirement verification, conflict resolution, and environmental adaptability verification.

[0240] Inertia requirement verification involves comparing the current required inertia with the rated inertia required for the current frequency band task. If the current required inertia is greater than or equal to a preset percentage of the rated inertia (e.g., high-frequency tasks require the current required inertia to be ≥90% of the rated inertia), it is determined that the requirement is met. If the current required inertia is less than a preset percentage of the rated inertia, it is determined that the requirement is not met.

[0241] The conflict handling mechanism is to prioritize adjusting the energy storage frequency regulation action (such as reducing power demand) to avoid conflicts if conflicts exist.

[0242] Environmental adaptability verification is performed in high-altitude weak power grid scenarios (referencing lifetime compensation factor). If the lifetime compensation factor is less than the minimum value of the preset lifetime compensation threshold range, it is determined that the environment exceeds the limit, and the current frequency band task type is downgraded to the next lower frequency band task.

[0243] Example:

[0244] Current frequency band task: High-frequency modulation task;

[0245] Inertia requirement: Current required inertia = 5MW⋅s (rated inertia = 5.5MW⋅s) → Requirement met;

[0246] Conflict detection: Conflict exists → Reduce energy storage frequency regulation power requirement to 10kW;

[0247] Environmental parameters: Altitude = 3500m, Temperature = 42℃, Lifetime compensation factor = 0.83, Minimum value of lifetime compensation threshold range = 0.85 → Forced downgrade to medium frequency task;

[0248] For handling abnormal scenarios, if there is insufficient inertia, the backup energy storage unit is activated (based on the backup device activation logic).

[0249] If the conflict cannot be resolved, that is, if the photovoltaic reactive power command conflicts with the energy storage frequency regulation action and the energy storage frequency regulation action cannot be adjusted, the energy storage frequency regulation action is suspended, and only photovoltaic reactive power compensation is performed (based on the power margin threshold).

[0250] If the environment exceeds the limit, switch completely to the next lower frequency band (based on the low-frequency task allocation strategy).

[0251] The aim is to resolve the conflict between photovoltaic reactive power commands and energy storage frequency regulation actions through dynamic responsibility allocation and task priority adjustment, and to generate executable performance commands.

[0252] Based on the current frequency change rate and the current frequency band task type matched from the three-dimensional performance mapping table, responsibilities are allocated by judging responsibility priority logic;

[0253] The logic for prioritizing responsibility is defined as follows: for high-frequency frequency regulation tasks, frequency regulation is prioritized through the relevant equipment of the energy storage unit (because of its fast response speed); for low-frequency frequency regulation tasks, reactive power compensation is prioritized through the relevant equipment of the photovoltaic unit (because of its stable regulation capability).

[0254] For medium-frequency regulation tasks, the energy storage unit and photovoltaic unit related equipment are combined, with the energy storage unit related equipment regulating the active power and the photovoltaic unit related equipment synchronously regulating the reactive power.

[0255] Real-time monitoring of the performance of energy storage unit and photovoltaic unit related equipment; if the energy storage unit related equipment is not responding well (e.g., SOC is too low), the active power regulation responsibility that the energy storage unit related equipment cannot handle is transferred to the photovoltaic unit related equipment through reactive power-active power coupling control.

[0256] Example:

[0257] If the task is a high-frequency frequency regulation task, energy storage should prioritize power adjustment, followed by photovoltaic reactive power commands.

[0258] If the matching is a low-frequency frequency regulation task, the photovoltaic system will prioritize reactive power adjustment, followed by the energy storage system's frequency regulation action.

[0259] If the matching is a medium-frequency frequency regulation task → energy storage adjusts power, and photovoltaic synchronous adjustment adjusts reactive power;

[0260] Responsibility transfer strategy for medium-frequency modulation tasks:

[0261] Active transfer: If the energy storage response speed is insufficient, actively transfer part of the frequency regulation responsibility to photovoltaic (indirectly stabilize the frequency through reactive power regulation).

[0262] Passive transfer: If the reactive power capacity of photovoltaic power is limited (such as inverter overload), the responsibility for reactive power regulation is forced to be transferred to energy storage (by adjusting the SOC to indirectly compensate the voltage).

[0263] Based on the responsibility allocation results, basic instructions are generated, including reactive power compensation instructions and frequency division power instructions;

[0264] The reactive power compensation command is calculated by taking the voltage change and lifetime compensation factor, combined with the equipment's maximum reactive power capacity (the rated reactive power capacity of the equipment (such as photovoltaic inverters or SVG)) and voltage sensitivity coefficient (calibrated according to grid characteristics), and then using this as the reactive power adjustment command.

[0265] The method for calculating reactive power adjustment is as follows: calculate the ratio of voltage change to the rated voltage of the power grid, and multiply the ratio by the voltage sensitivity coefficient, the maximum reactive power capacity of the equipment, and (1 - lifespan compensation factor) to obtain the reactive power adjustment.

[0266] The responsibility allocation logic is as follows: If the responsibility is transferred to the photovoltaic unit, based on the voltage change calculated by the AVC system (used to describe the voltage fluctuation or adjustment range in the power system; in medium frequency tasks, the voltage change represents the voltage change caused by the reactive power that the photovoltaic power station needs to adjust (e.g., ±5%)), a reactive power increment (e.g., +1.2MVar) is generated through reactive power compensation instructions and directly sent to the relevant equipment of the photovoltaic unit (e.g., photovoltaic inverters); if the responsibility is transferred to the energy storage unit, the priority of the photovoltaic reactive power instructions is reduced (e.g., only maintaining basic reactive power output), and voltage compensation is prioritized through the energy storage system (e.g., SVG or capacitor bank);

[0267] Application scenario example:

[0268] Industrial power application scenario: When the voltage change is -4% (voltage is low), the photovoltaic inverter needs to quickly output a reactive power adjustment of +0.72MVar.

[0269] In high-altitude scenarios: if the lifespan compensation factor is 0.85 (equipment lifespan degradation), then the reactive power adjustment needs to be multiplied by (1-0.8)=0.15, which significantly reduces the reactive power command of photovoltaic systems.

[0270] The frequency division power command is an active power adjustment amount calculated based on the frequency change rate and lifetime compensation factor, combined with the maximum power of the equipment (the rated active power of the equipment, such as energy storage systems or photovoltaic inverters) and the frequency response sensitivity coefficient (obtained through grid stability requirements).

[0271] The method for calculating the active power adjustment is as follows: multiply the frequency response sensitivity coefficient (obtained through grid stability requirement calibration) by the frequency change rate, the maximum power of the equipment, and the lifetime compensation factor to obtain the active power adjustment.

[0272] The responsibility allocation logic is as follows: If the responsibility is transferred to the energy storage unit, the energy storage frequency regulation power is calculated based on the frequency change rate and response delay parameters (the adjustment amount of power demand must meet the dynamic inertia constraint, that is, the current required inertia is less than or equal to the minimum safe inertia; otherwise, the active power adjustment amount needs to be corrected through the formula); if the responsibility is transferred to photovoltaic, the priority of the energy storage frequency regulation command is reduced (e.g., only maintaining SOC stability), and active power is adjusted first through the photovoltaic unit's related equipment (e.g., photovoltaic inverters must have fast active power regulation capability (e.g., response time ≤ 100ms)).

[0273] Application scenario example:

[0274] In the grid frequency regulation scenario: when the frequency change rate is 0.6Hz / s (high frequency fluctuation), the energy storage system needs to output an active power adjustment of 7.65kW (considering K_life=0.85).

[0275] High-temperature scenarios: If the lifespan compensation factor = 0.85 (equipment lifespan degradation), then the active power adjustment amount needs to be multiplied by 0.85 to reduce the energy storage frequency regulation power;

[0276] The basic instructions are compensated and modified based on environmental conditions and lifespan. The modification logic is as follows:

[0277] For high-altitude and high-temperature scenarios, if high altitude (e.g., >3000m) or high temperature (e.g., >40℃) is detected, the energy storage frequency regulation power is reduced proportionally according to the lifetime compensation factor (e.g., <0.85) (e.g., the energy storage frequency regulation power is multiplied by the lifetime compensation factor to obtain the reduced energy storage frequency regulation power).

[0278] Based on the device's classification score, if the device is classified as a low-frequency response cluster, the current frequency band task type will be forcibly downgraded (e.g., from high frequency to medium frequency), and the frequency modulation threshold will be adjusted proportionally (e.g., ±0.1Hz → ±0.15Hz).

[0279] In high-altitude and high-temperature scenarios, the dynamic inertia constraint is further integrated, and the corrected logic is as follows:

[0280] If the current required inertia is greater than the minimum safe inertia, reduce the energy storage frequency regulation power proportionally (the example reduction method is to first calculate the ratio of the minimum safe inertia to the current required inertia, and then multiply the calculation result by the energy storage frequency regulation power).

[0281] If the reactive power compensation command and the frequency division power command conflict in direction (e.g., reactive power compensation command = +1.2MVar vs frequency division power command = -20kW), the reactive power compensation command shall be given priority and the frequency division power weight shall be dynamically adjusted.

[0282] The final reactive power compensation command and frequency division power command are integrated to generate an efficiency command set, and the efficiency command set is defined to meet the constraints of reactive power compensation priority, frequency modulation power constraint and dynamic inertia guarantee.

[0283] When the voltage change is large (e.g., >±5%), the priority of the photovoltaic reactive power compensation command is forcibly increased.

[0284] The frequency regulation power constraint is that the energy storage frequency regulation power of the equipment must not exceed the maximum power of the equipment and the value after correction by the life compensation factor.

[0285] Dynamic inertia protection ensures that the current required inertia is less than or equal to the minimum safe inertia.

[0286] By simulating typical disturbances in high-altitude weak power grid scenarios and combining historical data to pinpoint the sources of deviations, the efficiency weight correction factor is dynamically adjusted.

[0287] Based on the performance instruction set, a simulation model is constructed according to high-altitude environmental parameters (e.g., altitude ≥ 3000m), weak grid characteristics (e.g., short-circuit capacity ratio ≤ 10%), and typical disturbance types (e.g., voltage drop, frequency fluctuation). The model simulates different types of typical disturbances (e.g., voltage drop (e.g., voltage change = -10% of rated voltage), frequency fluctuation (e.g., frequency change = 0.8Hz / s)) and injects historical data (e.g., equipment lifespan degradation records, execution effects of preceding reactive power compensation instructions and frequency division power instructions).

[0288] In the simulation environment, reactive power compensation commands and frequency division power commands are executed, and the actual response (such as voltage recovery time and frequency adjustment accuracy) is recorded; then the response deviation (such as the difference between the actual voltage change recovery value and the target value, and the frequency change rate adjustment error) is calculated.

[0289] By combining response deviations with historical data, regression analysis is used to pinpoint the main causes of the deviations (such as the life compensation factor not accurately reflecting the aging rate of high-altitude equipment), and the efficiency weighting factor is dynamically adjusted (such as adjusting the voltage sensitivity coefficient from 1.2 to 1.5 to compensate for the reactive power response lag of high-altitude equipment), including the frequency response sensitivity coefficient and the voltage sensitivity coefficient.

[0290] Based on the revised performance weighting factor, the performance score of each device is calculated using the performance scoring formula.

[0291] ;in, Represents the voltage sensitivity coefficient. Indicates the maximum reactive power capacity of the equipment. Indicates the lifespan compensation factor. Represents the frequency response sensitivity coefficient. Indicates the maximum power of the device;

[0292] The responsibility allocation rules are updated based on the equipment's performance rating (e.g., sorted by priority: energy storage unit > photovoltaic unit > traditional capacitor bank, and then sorted by performance rating from highest to lowest).

[0293] Set a performance score threshold, mark devices with performance scores below the threshold as inefficient devices, mark clusters with a certain proportion of inefficient devices as inefficient clusters, disband inefficient clusters and remove responsibility assignments.

[0294] Devices with performance scores greater than or equal to the performance score threshold are marked as high-efficiency devices, and a list of high-efficiency devices is generated.

[0295] Based on the list of efficient equipment, clusters are reorganized according to the geographical location of the equipment and the complementarity of their functions. The cluster combination is optimized through dynamic clustering algorithms (such as K-means) to ensure responsiveness.

[0296] Example 2

[0297] Please see Figure 3 As shown, parts not described in detail in this embodiment are described in Embodiment 1. A new energy grid-connected optimization control system based on an energy storage system is provided, including:

[0298] Dynamic Correction and Responsibility Generation Module: Deploy a sensor network to collect real-time data on grid impedance fluctuation amplitude, harmonic distortion rate, and ambient temperature, and obtain the initial SOC value of the energy storage unit. Correct the SOC value through the SOC estimation error correction mechanism, and then calculate the adjustable power margin and response delay parameters of the photovoltaic unit. Combine the lifetime degradation coefficient and temperature compensation factor to generate an efficiency responsibility table that includes reactive power responsibility zoning, frequency regulation threshold, and lifetime compensation factor.

[0299] Response cluster construction and 3D mapping generation module: Based on the performance responsibility table, the response cluster is constructed according to the adjustable power margin and response delay parameters. The frequency modulation threshold is matched with the response cluster type to generate frequency modulation task packages. The task execution rights are allocated through performance competition indicators to generate a time-frequency-power 3D performance mapping table.

[0300] Dynamic inertia calculation and cascade verification module: Based on the three-dimensional performance mapping table and performance responsibility table, combined with the grid frequency change rate, the inertia requirement is dynamically calculated, the conflict between photovoltaic reactive power command and energy storage frequency regulation action is detected, and cascade verification is performed; in high-altitude and high-temperature scenarios, the responsibility transfer strategy is automatically activated to generate a performance command set that integrates reactive power compensation, frequency division power and dynamic inertia.

[0301] Disturbance Analysis and Correction Factor Adjustment Module: Based on the performance instruction set, it analyzes response deviations by simulating typical disturbances in high-altitude weak power grid scenarios, locates the source of deviations by combining historical data, and dynamically adjusts the performance weight correction factor.

[0302] Rule update and cluster reorganization module: Based on the performance weight correction factor, update the responsibility allocation rules of the performance responsibility table, disband inefficient clusters, and reorganize highly adaptable units.

[0303] Example 3

[0304] This embodiment discloses an electronic 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 operation mode of the new energy grid-connected optimization control method based on the energy storage system described above.

[0305] Since the electronic device described in this embodiment is the electronic device used to implement the new energy grid-connected optimization control method based on energy storage system in the embodiments of this application, those skilled in the art can understand the specific implementation method and various variations of the electronic device in this embodiment based on the new energy grid-connected optimization control method based on energy storage system described in the embodiments of this application. Therefore, how the electronic device implements the method in the embodiments of this application will not be described in detail here. Any electronic device used by those skilled in the art to implement the new energy grid-connected optimization control method based on energy storage system in the embodiments of this application falls within the scope of protection of this application.

[0306] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0307] The above description is merely a preferred embodiment of the present invention, and the scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for users of ordinary technical skills, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A new energy grid-connected optimization control method based on an energy storage system, characterized in that, include: S1: Deploy a sensor network to collect real-time data on grid impedance fluctuation amplitude, harmonic distortion rate, and ambient temperature, and obtain the initial SOC value of the energy storage unit. Correct the SOC value through the SOC estimation error correction mechanism, and then calculate the adjustable power margin and response delay parameters of the photovoltaic unit. Combine the lifetime decay coefficient and temperature compensation factor to generate an efficiency responsibility table that includes reactive power responsibility zoning, frequency regulation threshold, and lifetime compensation factor. S2: Based on the performance responsibility table, construct a response cluster according to the adjustable power margin and response delay parameters, match the frequency modulation threshold with the response cluster type, generate frequency modulation task packages, allocate task execution rights through performance competition indicators, and generate a three-dimensional performance mapping table of time-frequency band-power. S3: Based on the three-dimensional performance mapping table and performance responsibility table, and combined with the grid frequency change rate, dynamically calculate the inertia requirement, detect the conflict between photovoltaic reactive power instructions and energy storage frequency regulation actions, and perform cascade verification; automatically activate the responsibility transfer strategy in high-altitude and high-temperature scenarios to generate a performance instruction set that integrates reactive power compensation, frequency division power and dynamic inertia; S4: Based on the performance instruction set, the response deviation is analyzed by simulating typical disturbances in high-altitude weak power grid scenarios, and the source of deviation is located by combining historical data, and the performance weight correction factor is dynamically adjusted. S5: Based on the performance weight correction factor, update the responsibility allocation rules of the performance responsibility table, disband inefficient clusters, and reorganize highly adaptable units. 2.The method of claim 1, wherein, The methods for correcting the SOC value include: By deploying a sensor network in the photovoltaic and energy storage area through a hybrid layout scheme, the grid impedance fluctuation amplitude, harmonic distortion rate, irradiance intensity and ambient temperature data of the photovoltaic units can be collected in real time. The SOC value of the energy storage unit is obtained from the battery management system as the initial SOC value; The collected data on power grid impedance fluctuation amplitude, harmonic distortion rate, ambient temperature, irradiance intensity, and initial SOC value were cleaned and aligned according to timestamps. Set a normal temperature threshold. If the ambient temperature data is greater than or equal to the normal temperature threshold, use the initial SOC value directly as the SOC value. If the ambient temperature data is lower than the normal temperature threshold, it is determined to be a low temperature environment, triggering the SOC estimation error correction mechanism to correct the initial SOC estimation value and obtain the corrected SOC value. The SOC estimation error correction mechanism is as follows: Define a reference temperature, and calculate the temperature compensation factor by combining ambient temperature data with the reference temperature. The product of the temperature compensation factor and the initial SOC estimate is taken as the corrected SOC value. 3.The method of claim 2, wherein, The calculation methods for the adjustable power margin and response delay parameters include: Set the normal range of SOC value and the default power suppression coefficient of adjustable power margin. Adjust the default power suppression coefficient according to the power regulation strategy based on the SOC value to obtain the adjusted power suppression coefficient. The logic for adjusting the SOC value and power suppression coefficient is as follows: If the SOC value is less than the minimum value of the normal range, the energy storage unit is determined to be in a low-charge state, and the power suppression coefficient is increased by a preset ratio; if the SOC value is greater than the maximum value of the normal range, the energy storage unit is determined to be in a high-charge state, and the power suppression coefficient is decreased by a preset ratio; if the SOC value is within the normal range, the default power suppression coefficient is maintained. Based on the irradiance and ambient temperature, combined with the module efficiency and module area of ​​the photovoltaic unit, the theoretical maximum power of the photovoltaic unit under the current environment is calculated. The grid harmonic limit for adjustable power margin is defined as setting a harmonic distortion threshold. If the harmonic distortion rate is less than or equal to the harmonic distortion threshold, the grid harmonic pollution is determined to be low, and the calculated theoretical maximum power remains unchanged. If the harmonic distortion rate is greater than the harmonic distortion threshold, the power grid is judged to have high harmonic pollution, and the calculated theoretical maximum power is reduced by a preset ratio. The instantaneous rate of change of photovoltaic unit power output is obtained as the power change rate. The range of adjustable power is dynamically adjusted according to the power change rate. Then, combined with the theoretical maximum power and power suppression coefficient obtained through grid harmonic limitation, the adjustable power margin is calculated. Based on the amplitude of grid impedance fluctuation and ambient temperature, if the amplitude of grid impedance fluctuation is greater than the preset amplitude threshold, the grid is determined to be unstable. An impedance fluctuation correction factor is calculated using the amplitude of grid impedance fluctuation to correct the reference response time. When the difference between the ambient temperature and the reference temperature is greater than the preset temperature deviation threshold, it is determined that the ambient temperature deviates from the reference temperature, and the temperature compensation factor is used to correct the reference response time. The impedance fluctuation correction factor and temperature compensation factor are multiplied and superimposed on the reference response time to obtain the final response delay parameter.

4. The energy storage system based new energy grid-connected optimization control method according to claim 3, characterized in that, The performance accountability table is generated in the following ways: Set the lifetime degradation coefficient of the energy storage unit. Based on the SOC value and ambient temperature data, correct the lifetime degradation coefficient according to the normal range of SOC and the normal temperature threshold. Take the product of the corrected lifetime degradation coefficient and the temperature compensation factor as the lifetime compensation factor. Based on adjustable power margin, SOC value and response delay parameters, and according to the characteristics of high-altitude weak power grid environment, combined with power grid harmonic distortion rate and ambient temperature, the complete photovoltaic-storage area is divided into different types of sub-regions, and the reactive power responsibility of related equipment of photovoltaic unit and energy storage unit is dynamically allocated to obtain reactive power responsibility zoning. Meanwhile, based on the SOC value and lifetime compensation factor, combined with adjustable power margin and response delay parameters, combined with threshold dynamic adjustment logic and different types of sub-region adaptation strategies, the frequency modulation threshold range of photovoltaic units and energy storage units is dynamically adjusted, thereby generating a dynamic frequency modulation threshold table. The reactive power responsibility zoning, dynamic frequency regulation threshold table, and lifetime compensation factor are integrated to generate the final performance responsibility table.

5. The energy storage system based new energy grid-connected optimal control method according to claim 4, characterized in that, The frequency modulation task packet is generated in the following ways: Based on the performance responsibility table, classification weights are set for adjustable power margin and response delay parameters using expert experience method, and then the classification score of the equipment is calculated using the classification score calculation formula. Set a score threshold range and classify devices whose classification scores are greater than or equal to the maximum value of the score threshold range into high-frequency response clusters; Devices whose classification scores are greater than or equal to the minimum value of the score threshold range but less than the score threshold range are classified as medium-frequency response clusters; Devices with classification scores less than the minimum value of the score threshold range are classified into low-frequency response clusters; Based on the divided response cluster types, according to the characteristics of different types of response clusters, the frequency modulation threshold and response cluster type are matched according to the task allocation logic to generate frequency modulation task packages for the corresponding frequency bands, including high-frequency frequency modulation tasks, medium-frequency frequency modulation tasks and low-frequency frequency modulation tasks. The task allocation logic is defined as follows: Real-time monitoring of power grid frequency change rate; setting a change rate threshold range; tasks with a frequency change rate greater than the maximum value of the change rate threshold range are designated as high-frequency frequency regulation tasks and are prioritized for execution by the high-frequency response cluster. Tasks whose frequency change rate is greater than or equal to the minimum value of the change rate threshold interval and less than the maximum value of the change rate threshold interval are designated as medium-frequency sub-frequency modulation tasks and executed by calling the medium-frequency response cluster. Tasks whose frequency change rate is less than the minimum value of the change rate threshold range are treated as low-frequency frequency modulation tasks and executed by calling the low-frequency response cluster. Define the adjustable power margin range, and divide the adjustable power margin into three levels according to the margin range, each corresponding to a different type of response cluster. Tasks that meet the corresponding frequency band but have a power requirement less than a certain percentage of the average power requirement of the corresponding frequency band are marked as hybrid task packages. If a device's classification score matches the classification of the response cluster, but its corresponding adjustable power margin does not match the corresponding response cluster classification, then the device will be marked as a degraded device and will execute a hybrid task in the corresponding response cluster. 6.The method of claim 5, wherein, The three-dimensional performance mapping table is generated in the following ways: Performance competitiveness metrics include response delay parameters, adjustable power margin, and lifetime compensation factor. The logic for allocating task execution rights is defined as follows: By assigning weights to each performance competitiveness indicator using expert experience, and then weighting and integrating these indicators, the overall competitiveness score of the equipment is obtained. Then, the comprehensive competition score is divided into three levels by threshold segmentation method, corresponding to high frequency frequency modulation tasks, medium frequency frequency modulation tasks and low frequency frequency modulation tasks respectively. Combined with the life compensation factor, the task execution rights of different frequencies are dynamically allocated to the equipment with the corresponding comprehensive competition score. Set a fixed time window, record the timestamp of task triggering according to the time window as the time dimension, take the frequency band type of the task as the frequency band dimension, take the three levels of adjustable power margin as the power dimension, integrate the time dimension, frequency band dimension and power dimension to construct a three-dimensional performance mapping table.

7. The energy storage system based new energy grid-connected optimization control method according to claim 6, characterized in that, The methods for calculating inertia requirements, detecting conflicts between photovoltaic reactive power commands and energy storage frequency regulation actions, and performing cascade verification include: Based on the current frequency change rate of the power grid, the current frequency band task type is matched from the three-dimensional performance mapping table, and the response delay parameters of the corresponding equipment and the frequency modulation threshold of the current frequency band task are extracted. The required inertia is then calculated using the inertia requirement formula. The current required inertia is compared with the preset minimum safe inertia. If the current required inertia is greater than or equal to the minimum safe inertia, it is marked as sufficient inertia. If the current required inertia is less than the minimum safe inertia, it is determined that the inertia is insufficient and the emergency frequency adjustment strategy is triggered. The method for detecting conflicts between photovoltaic reactive power commands and energy storage frequency regulation actions is as follows: Obtain the current photovoltaic reactive power command and energy storage demand frequency regulation action, and define the conflict condition between the photovoltaic reactive power command and energy storage demand frequency regulation action as follows: The photovoltaic reactive power command requires an increase in reactive power, but the energy storage frequency regulation action requires a decrease in active power; the photovoltaic reactive power command requires a decrease in reactive power, but the energy storage frequency regulation action requires an increase in active power. Compare the sign of the photovoltaic reactive power command with the direction of the energy storage frequency regulation action. If the sign and direction are opposite, it is determined to be a conflict; if the sign and direction are not opposite, it is determined to be a non-conflict. The logic for performing cascaded verification includes inertia requirement verification, conflict resolution, and environmental adaptability verification. Inertia requirement verification involves comparing the current required inertia with the rated inertia required for the current frequency band task. If the current required inertia is greater than or equal to a preset proportion of the rated inertia, it is determined that the requirement is met; if the current required inertia is less than a preset proportion of the rated inertia, it is determined that the requirement is not met. Conflict handling involves prioritizing adjustments to energy storage frequency regulation to avoid conflicts if they occur. Environmental adaptability verification is performed in high-altitude weak power grid scenarios. If the lifetime compensation factor is less than the minimum value of the preset lifetime compensation threshold range, it is determined that the environment exceeds the limit, and the current frequency band task type is downgraded to the next lower frequency band task.

8. The new energy grid-connected optimization control method based on energy storage system according to claim 7, characterized in that, The generation methods of the performance instruction set include: Based on the current frequency change rate and the current frequency band task type matched from the three-dimensional performance mapping table, responsibilities are allocated by judging responsibility priority logic; The logic for prioritizing responsibility is defined as follows: for high-frequency frequency regulation tasks, frequency regulation is prioritized through the relevant equipment of the energy storage unit; for low-frequency frequency regulation tasks, reactive power compensation is prioritized through the relevant equipment of the photovoltaic unit. For medium-frequency regulation tasks, the energy storage unit and photovoltaic unit related equipment are combined, with the energy storage unit related equipment regulating the active power and the photovoltaic unit related equipment synchronously regulating the reactive power. Real-time monitoring of the performance of energy storage unit and photovoltaic unit related equipment; if the energy storage unit related equipment is not responding well, the active power regulation responsibility that the energy storage unit related equipment cannot handle is transferred to the photovoltaic unit related equipment through reactive power-active power coupling control. Based on the responsibility allocation results, basic instructions are generated, including reactive power compensation instructions and frequency division power instructions; The reactive power compensation command is calculated by combining the voltage change and lifetime compensation factor with the equipment's maximum reactive power capacity and voltage sensitivity coefficient to determine the reactive power adjustment amount, which is then used as the reactive power compensation command. The responsibility allocation logic is as follows: if the responsibility is transferred to the photovoltaic unit, the reactive power increment is generated through reactive power compensation instructions based on the voltage change calculated by the AVC system and directly sent to the relevant equipment of the photovoltaic unit; if the responsibility is transferred to the energy storage unit, the priority of the photovoltaic reactive power instructions is reduced, and the voltage is compensated through the energy storage system first. The frequency division power command is based on the frequency change rate and lifetime compensation factor, combined with the equipment's maximum power calculation and the active power adjustment amount of the frequency response sensitivity coefficient. The responsibility allocation logic is as follows: if the responsibility is transferred to the energy storage unit, the energy storage frequency regulation power is calculated based on the frequency change rate and response delay parameters; if the responsibility is transferred to the photovoltaic unit, the priority of the energy storage frequency regulation command is reduced, and the active power is adjusted through the relevant equipment of the photovoltaic unit first. The basic instructions are compensated and modified based on environmental conditions and lifespan. The modification logic is as follows: For high-altitude and high-temperature scenarios, if high altitude is detected, the energy storage frequency regulation power will be reduced proportionally based on the lifespan compensation factor. Based on the device's classification score, if the device is classified as a low-frequency response cluster, the current frequency band task type will be forcibly downgraded, and the frequency modulation threshold will be adjusted proportionally. In high-altitude and high-temperature scenarios, the dynamic inertia constraint is further integrated, and the corrected logic is as follows: If the current required inertia is greater than the minimum safe inertia, reduce the energy storage frequency regulation power proportionally. If the reactive power compensation command and the frequency division power command conflict in direction, the reactive power compensation command shall be given priority and the frequency division power weight shall be dynamically adjusted. The constraints for defining the performance instruction set are reactive power compensation priority, frequency modulation power constraint, and dynamic inertia guarantee. The final reactive power compensation command and frequency division power command are integrated to generate an efficiency command set that meets the constraints.

9. The new energy grid-connected optimization control method based on energy storage system according to claim 8, characterized in that, The methods for dynamically adjusting the performance weight correction factor include: Based on the performance instruction set, a simulation model is constructed according to high-altitude environmental parameters, weak power grid characteristics and typical disturbance types to simulate different types of typical disturbances; In the simulation environment, reactive power compensation commands and frequency division power commands are executed, the actual response is recorded, and then the response deviation is calculated. By combining response deviations with historical data, regression analysis is used to pinpoint the main causes of the deviations, and the performance weighting factors, including the frequency response sensitivity coefficient and the voltage sensitivity coefficient, are dynamically adjusted.

10. The new energy grid-connected optimization control method based on energy storage system according to claim 9, characterized in that, The methods for updating the performance responsibility table, disbanding inefficient clusters, and reorganizing highly adaptable units include: Based on the revised performance weighting factor, the performance score of each device is calculated using the performance scoring formula. The responsibility allocation rules are updated based on the equipment's performance rating. Set a performance score threshold, mark devices with performance scores below the threshold as inefficient devices, mark clusters with a certain proportion of inefficient devices as inefficient clusters, disband inefficient clusters and remove responsibility assignments. Devices with performance scores greater than or equal to the performance score threshold are marked as high-efficiency devices, and a list of high-efficiency devices is generated. Based on a list of efficient equipment, clusters are reorganized according to geographical proximity and functional complementarity of the equipment, and the cluster combination is optimized through dynamic clustering algorithms to ensure responsiveness.

Images