Intelligent control method and system for operating state of energy storage system

Abstract

The application provides an intelligent control method and system for the operation state of an energy storage system, and belongs to the field of operation state control. Real-time operation data and historical power data of the energy storage system are obtained to construct a power reference trajectory expected to be exchanged at a grid connection point of the energy storage system. Dynamic feedback correction amounts of the states of each energy storage unit are generated based on the real-time operation data, and the power reference trajectory is corrected according to all the dynamic feedback correction amounts to obtain a corrected power reference trajectory. The optimal power control sequence of each energy storage unit in the energy storage system is determined through the corrected power reference trajectory, and the instructions of the optimal power control sequence are issued to the corresponding energy storage unit for execution based on an industrial cloud platform. The scheme of the application can intelligently control the operation state of the energy storage system based on the power time sequence evolution law of each energy storage unit and the differentiated energy storage response characteristics.

Description

Technical Field

[0001] This application relates to the field of operational status control technology, and more specifically, to an intelligent control method and system for the operational status of an energy storage system. Background Technology

[0002] Operational status control refers to a management mechanism that monitors, evaluates, and adjusts the real-time operating parameters of a system, equipment, or process to ensure its continuous operation within a preset range of safety, efficiency, and stability. It achieves closed-loop control from manual to fully automatic operation, aiming to optimize performance, extend lifespan, and mitigate operational risks.

[0003] Currently, traditional energy storage operation and control methods are relatively crude, mostly relying solely on instantaneous operating data to formulate scheduling schemes. They fail to combine historical power variation patterns and the response characteristics of different energy storage units to rationally plan grid-connected power trajectories. Furthermore, they fail to incorporate real-time state of charge of energy storage units and long-term health indicators of the system for dynamic state correction. Power regulation lacks a closed-loop optimization mechanism, and the operational safety constraints of each energy storage unit are ignored during power allocation. This easily leads to problems such as unreasonable power allocation of energy storage units, abnormal charging and discharging behavior, and accelerated equipment aging. It is difficult to ensure smooth and orderly power interaction between the energy storage system and the grid connection point, and it also reduces the overall operating efficiency and service life of the energy storage system. It cannot meet the actual usage requirements of intelligent, refined, and long-term stable operation of energy storage systems in industrial scenarios. Therefore, how to intelligently control the operating status of energy storage systems based on the power time-series evolution patterns and differentiated energy storage response characteristics of each energy storage unit has become a problem facing the industry. Summary of the Invention

[0004] This application provides an intelligent control method and system for the operating status of an energy storage system, which can intelligently control the operating status of the energy storage system based on the power time-series evolution law and differentiated energy storage response characteristics of each energy storage unit.

[0005] In a first aspect, this application provides an intelligent control method for the operating status of an energy storage system, applied to an industrial cloud platform. The energy storage system includes at least two types of energy storage units with different response characteristics. The method includes the following steps: The industrial cloud platform is used to obtain real-time operating data and historical power data of the energy storage system. Based on the long-term power variation trend in historical power data and the response characteristics of each energy storage unit, a power reference trajectory is constructed for the expected power exchange between the energy storage system and the grid connection point. Based on the state of charge of each energy storage unit in the real-time operating data and the preset central value that characterizes the long-term healthy operating state of the energy storage system, a dynamic feedback correction amount for the state of each energy storage unit is generated. The power reference trajectory is then closed-loop corrected according to all the dynamic feedback correction amounts to obtain the corrected power reference trajectory. The power of the energy storage system is iteratively optimized by using the power safety constraints of each energy storage unit and the corrected power reference trajectory to obtain the optimal power control sequence for each energy storage unit in the energy storage system. The instructions of the optimal power control sequence are then sent to the corresponding energy storage unit for execution based on the industrial cloud platform.

[0006] In some embodiments, the real-time operating data includes the voltage, current, state of charge, temperature, and health status of each energy storage unit, and the historical power data includes the day-ahead, intraday, and weekly power time-series data of the energy storage system at the grid connection point.

[0007] In some embodiments, constructing a power reference trajectory for the expected exchange between the energy storage system and the grid connection point based on the long-term power variation trend in historical power data and the response characteristics of each energy storage unit specifically includes: Empirical mode decomposition is performed on the historical power data to extract its long-term trend components and short-term fluctuation components; Obtain the response characteristics of each energy storage unit; Based on the response characteristics of each energy storage unit, the long-term trend component is allocated to the energy storage unit with a slower response speed, and the short-term fluctuation component is allocated to the energy storage unit with a faster response speed. The weighted summation of each allocated component is performed, and a power reference trajectory for the expected exchange between the energy storage system and the grid connection point is constructed based on the grid dispatch instructions and the voltage deviation constraint at the grid connection point.

[0008] In some embodiments, generating a dynamic feedback correction amount for the state of each energy storage unit based on the state of charge of each energy storage unit in the real-time operating data and a preset central value characterizing the long-term healthy operating state of the energy storage system specifically includes: Obtain the safe operating range of the state of charge of each energy storage unit and the preset central value; The deviation between the state of charge of each energy storage unit and the preset central value is calculated to obtain the basic correction component of each energy storage unit; Based on the differences in response characteristics of each energy storage unit and the safe operating range of the state of charge, the basic correction components of each energy storage unit are matched in time scale and limited in amplitude to generate dynamic feedback correction quantities for the state of each energy storage unit.

[0009] In some embodiments, the power reference trajectory is closed-loop corrected based on all dynamic feedback correction values ​​to obtain the corrected power reference trajectory, specifically including: The total correction amount is obtained by vector summation of the dynamic feedback corrections of all energy storage units after time synchronization. The total correction amount is superimposed on the power reference trajectory to form a preliminary correction trajectory; By using the power change rate constraint at the grid connection point to limit the slope of the preliminary correction trajectory, a corrected power reference trajectory that meets the grid safety requirements is obtained.

[0010] In some embodiments, the power of the energy storage system is iteratively optimized using the power safety constraints of each energy storage unit and the corrected power reference trajectory to obtain the optimal power control sequence for each energy storage unit in the energy storage system. Specifically, this includes: A multi-objective function is established with the goal of minimizing tracking error, balancing the state of charge of each energy storage unit, and minimizing the overall lifespan loss of the energy storage system. The corrected power reference trajectory is used as the tracking target, and the upper and lower limits of power, the upper limit of power change rate, and the state of charge boundary of each energy storage unit are used as power safety constraints. Using the power safety constraints as the boundary, the multi-objective function is solved by rolling optimization. The response characteristic weight matrix of each energy storage unit is introduced, and the power allocation value in multiple time domains in the future is calculated iteratively. The iteration is terminated when the rate of change of the multi-objective function between two adjacent iterations is less than a preset threshold or the maximum number of iterations is reached. The optimal power control sequence of each energy storage unit in the control time domain is output.

[0011] In some embodiments, the process of issuing instructions for the optimal power control sequence to the corresponding energy storage unit based on the industrial cloud platform specifically includes: In the industrial cloud platform, virtual device objects are constructed that correspond one-to-one with each energy storage unit. The virtual device objects store the communication protocol, address mapping, and power response delay parameters of the corresponding energy storage unit. The optimal power control sequence is split into single-step instructions according to timestamps, and the instruction issuance time is pre-compensated according to the power response delay parameters of each energy storage unit. The pre-compensated instructions are sent to the local controller of the corresponding energy storage unit through the edge gateway of the industrial cloud platform, and the local controller drives the power conversion unit to execute them.

[0012] Secondly, this application provides an intelligent control system for the operating status of an energy storage system, applied to an industrial cloud platform. The energy storage system includes at least two types of energy storage units with different response characteristics. The system includes: The acquisition module is used to acquire real-time operating data and historical power data of the energy storage system through the industrial cloud platform; The processing module is used to construct the power reference trajectory that the energy storage system is expected to exchange at the grid connection point with the grid based on the long-term power variation trend in historical power data and the response characteristics of each energy storage unit. The processing module is also used to generate dynamic feedback correction quantities for the state of charge of each energy storage unit and the preset central value characterizing the long-term healthy operation state of the energy storage system based on the state of charge of each energy storage unit in the real-time operation data, and to perform closed-loop correction on the power reference trajectory based on all the dynamic feedback correction quantities to obtain the corrected power reference trajectory. The execution module is used to iteratively optimize the power of the energy storage system by using the power safety constraints of each energy storage unit and the corrected power reference trajectory to obtain the optimal power control sequence of each energy storage unit in the energy storage system, and to issue the instructions of the optimal power control sequence to the corresponding energy storage unit for execution based on the industrial cloud platform.

[0013] Thirdly, this application provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described intelligent control method for the operating state of the energy storage system.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described intelligent control method for the operating state of an energy storage system.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: The intelligent control method and system for the operating status of the energy storage system provided in this application acquires real-time operating data and historical power data through an industrial cloud platform. This not only captures the power time-series evolution pattern but also grasps the current operating status of each energy storage unit. A power reference trajectory is constructed based on the long-term historical power variation trend and the response characteristics of each energy storage unit. This ensures that the reference trajectory not only conforms to the system's power time-series evolution pattern but also adapts to the response differences of different energy storage units, avoiding the problem of the reference trajectory being out of sync with actual operating requirements. Dynamic feedback correction quantities are generated based on the state of charge of each energy storage unit and the preset central value for long-term healthy operation of the system. This performs closed-loop correction on the power reference trajectory, achieving real-time adaptation to power time-series changes and differentiated states of energy storage units. This improves the accuracy of the reference trajectory, avoids control deviations caused by fluctuations in the state of energy storage units, and ensures the long-term healthy operation of the system. Power iterative optimization is performed through power safety constraints of each energy storage unit and the corrected reference trajectory to generate the optimal power control sequence for each energy storage unit and issue it for execution. This ensures that the control command for each energy storage unit matches its response characteristics and follows the power time-series evolution pattern, achieving intelligent regulation of the energy storage system's operating status. By adopting the scheme of this application, the operating status of the energy storage system can be intelligently controlled based on the power time-series evolution law and differentiated energy storage response characteristics of each energy storage unit. Attached Figure Description

[0016] Figure 1 This is an exemplary flowchart of an intelligent control method for the operating state of an energy storage system according to some embodiments of this application; Figure 2 This is an exemplary flowchart illustrating the determination of a power reference trajectory according to some embodiments of this application; Figure 3 This is a flowchart illustrating the closed-loop intelligent control process of an energy storage system according to some embodiments of this application; Figure 4 This is a schematic diagram of the structure of an intelligent control system for the operation status of an energy storage system according to some embodiments of this application; Figure 5 This is a schematic diagram of the structure of a computer device for implementing an intelligent control method for the operating state of an energy storage system, according to some embodiments of this application. Detailed Implementation

[0017] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] refer to Figure 1 The figure is an exemplary flowchart of an intelligent control method for the operating state of an energy storage system according to some embodiments of this application. The intelligent control method for the operating state of the energy storage system mainly includes the following steps: In step 101, real-time operating data and historical power data of the energy storage system are obtained through the industrial cloud platform.

[0019] It should be noted that the real-time operating data in this application is a set of numerical information used to characterize the electrical state, thermal state and aging degree of each energy storage unit at each sampling moment. The real-time operating data includes the voltage, current, state of charge, temperature and health status of each energy storage unit. The historical power data is a time-series numerical sequence used to reflect the actual exchange power change pattern of the energy storage system grid connection point over multiple time scales in the past. The historical power data includes the day-ahead, intraday and weekly power time-series data of the energy storage system at the grid connection point.

[0020] In practice, the industrial cloud platform continuously collects voltage, current, state of charge, temperature, and health status values ​​from the battery management system and power conversion system of each energy storage unit at a fixed sampling period through a pre-established communication link. Simultaneously, it collects real-time power values ​​from the power quality monitoring device at the grid connection point. All the collected values ​​constitute the real-time operating data. The industrial cloud platform retrieves the grid connection point power values ​​stored in its internal time-series database for the past thirty days. These power values ​​are organized at different time resolutions: one point every fifteen minutes before the day, one point every five minutes within the day, and one point every hour on a weekly scale, forming historical power data. Outliers in the real-time operating data are replaced using linear interpolation of the values ​​from previous and subsequent times. Missing moments in the historical power data are filled with the average of adjacent complete periods. Different levels of health status values ​​are converted into standardized values ​​between 0 and 1. Finally, the processed real-time operating data and historical power data serve as the input basis for subsequent steps.

[0021] In step 102, a power reference trajectory for the expected exchange between the energy storage system and the grid connection point is constructed based on the long-term power variation trend in historical power data and the response characteristics of each energy storage unit.

[0022] In some embodiments, reference Figure 2 The figure is an exemplary flowchart of determining the power reference trajectory in some embodiments of this application. In this embodiment, the power reference trajectory to be exchanged between the energy storage system and the grid connection point can be constructed based on the long-term trend of power change in historical power data and the response characteristics of each energy storage unit. This can be achieved by the following steps: In step 1021, empirical mode decomposition is performed on the historical power data to extract its long-term trend component and short-term fluctuation component. In step 1022, the response characteristics of each energy storage unit are obtained; In step 1023, the long-term trend component is allocated to the energy storage unit with a slower response speed, and the short-term fluctuation component is allocated to the energy storage unit with a faster response speed, based on the response characteristics of each energy storage unit. In step 1024, the weighted superposition of each allocated component is performed, and based on the grid dispatch command and the voltage deviation constraint at the grid connection point, a power reference trajectory for the expected exchange between the energy storage system and the grid connection point is constructed.

[0023] It should be noted that the long-term trend component in this application is a time-series signal used to characterize the background component with the lowest frequency and slowest change in historical power data. The short-term fluctuation component is a time-series signal used to characterize the transient component with higher frequency and faster change in historical power data. The response characteristic is an inherent parameter used to describe how quickly each energy storage unit receives a power command and its actual output power reaches the commanded value. The allocated component is a time-series signal used to indicate the power baseline value that each energy storage unit should bear. The power reference trajectory is a target curve used to track the expected output power of the energy storage system at the grid connection point during future control cycles.

[0024] In specific implementation, empirical mode decomposition is performed on the historical power data to extract its long-term trend component and short-term fluctuation component. Specifically, the industrial cloud platform takes the historical power data as the input signal, first identifies all local maxima and local minima in the signal, fits the upper and lower envelopes respectively through cubic spline interpolation, calculates the mean sequence of the upper and lower envelopes, and subtracts the mean sequence from the original signal to obtain the first candidate component. The above operation is repeated until the first candidate component satisfies the two conditions of the intrinsic mode function, that is, the number of extreme points in the whole signal is equal to or differs from the number of zero-crossing points by at most one, and the local mean is zero at any time. At this time, the first candidate component is taken as an intrinsic mode function component, and the component is subtracted from the original signal to obtain the remaining signal. The above decomposition process is continued on the remaining signal to separate multiple intrinsic mode function components with frequencies from high to low in sequence. When the remaining signal presents a monotonic curve or the number of extreme points is less than two, the decomposition is stopped, the remaining signal is taken as the long-term trend component, and the sum of the first three intrinsic mode function components is taken as the short-term fluctuation component.

[0025] In specific implementation, the response characteristics of each energy storage unit are obtained as follows: The industrial cloud platform reads the rated power value, response time constant value, and charge / discharge conversion time value of each energy storage unit from the energy storage unit parameter configuration table pre-stored within the platform; where the response time constant is defined as the time required for the energy storage unit's step response to reach 63.2% of its final stable value, and the charge / discharge conversion time is defined as the time required for the energy storage unit to switch from its rated charging power to its rated discharging power; based on the read response time constant value, energy storage units with a response time constant of less than 100 milliseconds are marked as fast response units, and energy storage units with a response time constant of greater than or equal to one second are marked as slow response units, and the unit identifier of each energy storage unit and the marking result of its response characteristics are stored in a lookup table.

[0026] In specific implementation, the long-term trend component is allocated to energy storage units with slower response speeds, and the short-term fluctuation component is allocated to energy storage units with faster response speeds, based on the response characteristic labels of each energy storage unit in the lookup table. Specifically, the industrial cloud platform groups all energy storage units labeled as slow-response units into a slow-response unit set and all energy storage units labeled as fast-response units into a fast-response unit set, according to the response characteristic labels of each energy storage unit in the lookup table. The rated power of all energy storage units in the slow-response unit set is calculated, and then the proportion of the rated power of each slow-response unit to this sum is calculated. This proportion is then multiplied... The power base value that the slow response unit should bear at the current moment is obtained by using the value of the long-term trend component at the current moment. Similarly, the sum of the rated power of all energy storage units in the fast unit set is calculated, and the proportion of the rated power of each fast response unit to the sum is multiplied by the value of the short-term fluctuation component at the current moment to obtain the power base value that the fast response unit should bear at the current moment. For each energy storage unit, the power base value calculated above is expanded over time into a time series with the same time length as the long-term trend component and the short-term fluctuation component. This time series is the allocated component of the energy storage unit.

[0027] In specific implementation, the weighted summation of each allocated component, and based on the grid dispatch instructions and grid connection point voltage deviation constraints, a power reference trajectory for the expected exchange between the energy storage system and the grid connection point is constructed. Specifically, the industrial cloud platform receives the active power plan curve from the grid dispatch center. This curve provides the planned exchange power value of the grid connection point for the next four hours at a resolution of one point every fifteen minutes. Linear interpolation is used to improve its time resolution to the same one-second resolution as the allocated components. Simultaneously, the current effective voltage value of the grid connection point is read in real time from the power quality monitoring device at the grid connection point. The deviation between this voltage value and the rated voltage value is calculated, and this deviation is multiplied by a preset proportional coefficient of 0.1 to obtain the voltage correction power value. For each... Each energy storage unit is assigned a weighted component, which is multiplied by a preset unit weighting coefficient. The unit weighting coefficient is determined based on the unit's response characteristics and health status. The weighting coefficient for fast-response units is set to 0.8 by default, and the weighting coefficient for slow-response units is set to 0.2 by default. The weighted components of all energy storage units are summed at the same time to obtain the initial power trajectory. The initial power trajectory is summed with the grid dispatch command curve, and then the voltage correction power value is subtracted to obtain the preliminary power trajectory. The preliminary power trajectory is subjected to a first-order low-pass filter with a filtering time constant of two seconds to eliminate high-frequency glitches in the trajectory. The filtered trajectory is used as the power reference trajectory for the expected exchange between the energy storage system and the grid connection point.

[0028] In step 103, dynamic feedback correction values ​​for the state of charge of each energy storage unit in the real-time operating data and preset central values ​​characterizing the long-term healthy operating state of the energy storage system are generated. The power reference trajectory is then closed-loop corrected based on all the dynamic feedback correction values ​​to obtain the corrected power reference trajectory.

[0029] In some embodiments, generating the dynamic feedback correction amount for the state of each energy storage unit based on the state of charge of each energy storage unit in the real-time operating data and a preset central value characterizing the long-term healthy operating state of the energy storage system can be achieved through the following steps: Obtain the safe operating range of the state of charge of each energy storage unit and the preset central value; The deviation between the state of charge of each energy storage unit and the preset central value is calculated to obtain the basic correction component of each energy storage unit; Based on the differences in response characteristics of each energy storage unit and the safe operating range of the state of charge, the basic correction components of each energy storage unit are matched in time scale and limited in amplitude to generate dynamic feedback correction quantities for the state of each energy storage unit.

[0030] It should be noted that the safe operating range of state of charge (SOC) in this application is a numerical range used to define the lower and upper limits that the SOC of each energy storage unit should not exceed during normal operation. The preset central value is a fixed value used to indicate the ideal target point that the SOC of each energy storage unit should tend towards during long-term healthy operation. The basic correction component is the original power adjustment signal used to reflect the degree and direction of deviation of the current SOC of each energy storage unit from the preset central value. The dynamic feedback correction amount is a filtered and limited power increment signal used to perform closed-loop compensation of the power reference trajectory based on the current state of the energy storage unit.

[0031] In specific implementation, obtaining the safe operating range and preset central value of the state of charge (SOC) for each energy storage unit is as follows: The industrial cloud platform reads the lower limit, upper limit, and preset central value of the safe operating range of the SOC for each energy storage unit from the pre-stored energy storage unit parameter configuration table. The lower limit is set to 10%, the upper limit to 90%, and the preset central value to 50%. These values ​​are determined based on the SOC range with the lowest aging rate in the energy storage unit cycle life test data. For flywheel energy storage units, their SOC is equivalent to the ratio of the current rotor speed to the rated speed. The lower limit of the safe operating range of this ratio is set to 20%, the upper limit to 90%, and the preset central value to 50%. The lower limit and upper limit of the safe operating range of the SOC of all energy storage units are used as the range boundaries, and the preset central value is used as the target reference point.

[0032] In specific implementation, the deviation between the state of charge (SBC) of each energy storage unit and the preset central value is calculated to obtain the basic correction component for each energy storage unit. Specifically, the industrial cloud platform reads the current SBC value of each energy storage unit from real-time operating data; the current SBC value is subtracted from the preset central value of the energy storage unit to obtain a signed SBC deviation value, where a positive deviation value indicates that the current SBC is higher than the preset central value and requires discharge direction correction, and a negative deviation value indicates that the current SBC is lower than the preset central value and requires charging direction correction; the SBC deviation value is multiplied by a preset proportional coefficient of 0.05 to obtain the original power adjustment amount, which is the basic correction component; the above calculation is performed independently for each energy storage unit to obtain the basic correction component for each energy storage unit. The preset proportional coefficient is determined based on the ratio of the rated total power of the energy storage system to the maximum allowable range of the state of charge deviation, combined with the balance between tracking response speed and system stability during engineering commissioning. Specifically, when the state of charge deviation is 10%, the corresponding original power adjustment is approximately 0.5% of the rated total power of the energy storage system. This value ensures that sufficient and effective power correction is generated when the state of charge deviates from the central value, while avoiding excessive correction that could cause drastic fluctuations in the power reference trajectory and affect grid safety. Through multiple simulations and field tests, 0.05 has been verified to achieve fast convergence speed and small overshoot control under all typical operating conditions.

[0033] In practical implementation, the industrial cloud platform performs time-scale matching processing on the basic correction component of each energy storage unit based on its response characteristic markings. Specifically, for energy storage units marked as fast-response units, a first-order low-pass filter is applied to the basic correction component using a 50-millisecond time constant to match the rate of change of the correction amount with the actual power response capability of the fast-response unit; for energy storage units marked as slow-response units, a first-order low-pass filter is applied to the basic correction component using a 2-second time constant to match the rate of change of the correction amount with the power ramp-up capability of the slow-response unit. After completing the time-scale matching, the amplitude of the filtered correction component is then limited according to the safe operating range of the state of charge of each energy storage unit. Specifically, if the current state of charge is far from the safe operating range... If the lower limit of the operating range is less than 5% by 10%, the direction of the correction component is determined. If the direction of the correction component is the charging direction (i.e., negative), the amplitude of the correction component is linearly reduced proportionally to the distance from the boundary. The reduction factor is the current state of charge minus the lower limit and divided by 5%, while the direction is forced to point inward into the range (i.e., the discharging direction). If the current state of charge is less than 5% of the upper limit of the safe operating range (90%), and the direction of the correction component is the discharging direction (i.e., positive), the amplitude is also reduced proportionally and the direction is forced to point in the charging direction. If the current state of charge is in the middle of the safe operating range and more than 5% away from the boundary, no amplitude reduction is applied to the correction component. The correction component after time-scale matching and amplitude limiting is used as the dynamic feedback correction amount for the state of each energy storage unit.

[0034] In some embodiments, the power reference trajectory is closed-loop corrected based on all dynamic feedback corrections to obtain the corrected power reference trajectory, which can be achieved by the following steps: The total correction amount is obtained by vector summation of the dynamic feedback corrections of all energy storage units after time synchronization. The total correction amount is superimposed on the power reference trajectory to form a preliminary correction trajectory; By using the power change rate constraint at the grid connection point to limit the slope of the preliminary correction trajectory, a corrected power reference trajectory that meets the grid safety requirements is obtained.

[0035] It should be noted that the total correction amount in this application is the total power adjustment used to characterize the combined effect of the dynamic feedback correction amounts of all energy storage units within the same control cycle. The preliminary correction trajectory is the power tracking curve after compensation by the total correction amount, used as the input of the slope limiting stage. The corrected power reference trajectory is the final power curve that satisfies the power change rate constraint at the grid connection point, used as the tracking target of the power optimization allocation stage.

[0036] In specific implementation, the dynamic feedback correction values ​​of all energy storage units are synchronized over time and then vector-summed to obtain the total correction value. Specifically, at the beginning of each control cycle, the industrial cloud platform reads the power increment value corresponding to each energy storage unit in the current cycle from the dynamic feedback correction values ​​of each energy storage unit generated in the previous sub-step. Each dynamic feedback correction value has a positive and negative sign. A positive value indicates that the output power of the grid connection point needs to be increased, i.e., the discharge direction, and a negative value indicates that the output power of the grid connection point needs to be decreased, i.e., the charging direction. Since all dynamic feedback correction values ​​are calculated based on the same time base, the industrial cloud platform directly performs algebraic summation on the dynamic feedback correction values ​​of each energy storage unit, that is, adds all positive values ​​and all negative values ​​to obtain the net power adjustment value. This net power adjustment value is used as the total correction value.

[0037] In specific implementation, the total correction amount is superimposed on the power reference trajectory to form a preliminary correction trajectory. Specifically, the industrial cloud platform extracts the power reference value corresponding to the current moment of the control cycle from the power reference trajectory, adds the power reference value to the total correction amount obtained in the first sub-step, and obtains the preliminary corrected power value at that moment. The above superposition operation is repeated for each time point in the power reference trajectory, that is, the power reference value at each time point is added to the total correction amount of the same time point, where the total correction amount may be different at different time points. This generates a new curve with the same time length as the power reference trajectory, but the overall offset is determined by the total correction amount at each moment. This new curve is used as the preliminary correction trajectory.

[0038] In specific implementation, the slope of the preliminary correction trajectory is limited by the power change rate constraint at the grid connection point to obtain a corrected power reference trajectory that meets the grid safety requirements. Specifically, the industrial cloud platform reads the limit value of the power change rate constraint at the grid connection point from the grid safety operation procedures. This limit value stipulates that the change in active power at the grid connection point within one second shall not exceed 10% of the rated total power of the energy storage system. The preliminary correction trajectory is traversed point by point in chronological order. Starting from the second time point, the absolute value of the difference between the power value at the current time point and the power value at the previous time point is calculated. If the absolute value is less than or equal to the power change rate constraint limit value, the power value at the current time point remains unchanged. If the absolute value is greater than the power change rate constraint limit value, the sign of the difference between the power value at the current time point and the power value at the previous time point is calculated. This sign value is multiplied by the power change rate constraint limit value and then added to the power value at the previous time point to obtain the corrected power value at the current time point, thereby forcing the power change amount at this step to meet the constraint. After the above point-by-point limiting processing is completed for the entire preliminary correction trajectory, the limited trajectory is used as the corrected power reference trajectory.

[0039] In step 104, the power of the energy storage system is iteratively optimized by using the power safety constraints of each energy storage unit and the corrected power reference trajectory to obtain the optimal power control sequence of each energy storage unit in the energy storage system. The instructions of the optimal power control sequence are then sent to the corresponding energy storage units for execution based on the industrial cloud platform.

[0040] In some embodiments, the optimal power control sequence for each energy storage unit in the energy storage system can be obtained by iteratively optimizing the power of the energy storage system using the power safety constraints of each energy storage unit and the corrected power reference trajectory, which can be achieved through the following steps: A multi-objective function is established with the goal of minimizing tracking error, balancing the state of charge of each energy storage unit, and minimizing the overall lifespan loss of the energy storage system. The corrected power reference trajectory is used as the tracking target, and the upper and lower limits of power, the upper limit of power change rate, and the state of charge boundary of each energy storage unit are used as power safety constraints. Using the power safety constraints as the boundary, the multi-objective function is solved by rolling optimization. The response characteristic weight matrix of each energy storage unit is introduced, and the power allocation value in multiple time domains in the future is calculated iteratively. The iteration is terminated when the rate of change of the multi-objective function between two adjacent iterations is less than a preset threshold or the maximum number of iterations is reached. The optimal power control sequence of each energy storage unit in the control time domain is output.

[0041] It should be noted that the multi-objective function in this application is a weighted mathematical expression used to comprehensively quantify the merits of power allocation schemes across three dimensions: tracking accuracy, state-of-charge balance, and lifetime loss. The tracking objective is a standard reference curve used to measure the deviation between the total output power of the energy storage system and the desired value. Power safety constraints are a set of physical limits that each energy storage unit must not exceed during power allocation. Constraint boundaries are a set of inequality boundary conditions used to limit the range of values ​​for optimization variables. The response characteristic weight matrix is ​​a diagonal weight matrix used to apply different penalties to the power allocation value based on the response speed of each energy storage unit. The power allocation value is a variable to be solved, representing the active power output of each energy storage unit in each future control cycle. The optimal power control sequence is an ordered set of power allocation values ​​for each energy storage unit that minimizes the multi-objective function and satisfies all power safety constraints.

[0042] In specific implementation, a multi-objective function is established with the goals of minimizing tracking error, balancing the state of charge (SOC) of each energy storage unit, and minimizing the overall lifespan loss of the energy storage system. Specifically, the industrial cloud platform first defines a sub-objective to minimize tracking error, which is the square of the difference between the corrected power reference trajectory and the sum of the actual output power of all energy storage units. Then, a sub-objective to balance SOC is defined, which is the sum of the squares of the deviations between the real-time SOC of each energy storage unit and a preset central value (50%). Finally, a sub-objective to minimize the overall lifespan loss of the energy storage system is defined, which is the sum of the absolute values ​​of the power changes of each energy storage unit in adjacent control cycles. The industrial cloud platform assigns weight coefficients to these three sub-objectives: 0.5 for minimizing tracking error, 0.3 for SOC, and 0.2 for minimizing the overall lifespan loss of the energy storage system. The three sub-objectives are multiplied by their respective weight coefficients and then summed to obtain a total mathematical expression, which is used as the multi-objective function. The aforementioned weighting coefficients were determined based on the analytic hierarchy process (AHP) combined with expert experience. First, a judgment matrix was constructed using tracking accuracy, state of charge balance (SOP), and lifetime loss as criteria. Energy storage system operation and maintenance experts were invited to score the importance of each pair of parameters. The importance scale of tracking accuracy relative to SOP was 3:2, and its importance scale relative to lifetime loss was 5:2. The importance scale of SOP relative to lifetime loss was also 3:2. After normalization calculation and consistency verification, the allocation results were 0.5, 0.3, and 0.2. This weighting allocation can enable the system to prioritize meeting the tracking requirements of grid dispatch commands in simulation verification under different operating conditions, while also taking into account the lifetime protection and energy balance of energy storage units.

[0043] In specific implementation, the modified power reference trajectory is used as the tracking target, and the upper and lower limits of power, the upper limit of power change rate, and the state of charge boundary of each energy storage unit are used as power safety constraints. Specifically, the industrial cloud platform obtains the expected power value at each control moment from the modified power reference trajectory and uses the entire modified power reference trajectory as the tracking target. At the same time, the industrial cloud platform reads the maximum rated power of each energy storage unit from the energy storage unit parameter configuration table as the upper limit of power, the opposite of the maximum rated power as the lower limit of power, reads that the power change of each energy storage unit per second must not exceed 20% of the rated power as the upper limit of power change rate, and reads the lower limit of 10% and the upper limit of 90% of the state of charge of each energy storage unit as the state of charge boundary value. All the above values ​​are used together as power safety constraints.

[0044] In specific implementation, the power safety constraints are used as the constraint boundaries to perform rolling optimization on the multi-objective function. The response characteristic weight matrix of each energy storage unit is introduced, and the power allocation value in multiple future time domains is iteratively calculated. The iteration terminates when the rate of change of the multi-objective function between two adjacent iterations is less than a preset threshold or the maximum number of iterations is reached. The optimal power control sequence for each energy storage unit in the control time domain is output. Specifically, the industrial cloud platform uses the upper and lower limits of power, the upper limit of power change rate, and the state of charge boundary of each energy storage unit in the power safety constraints as insurmountable constraint boundaries for optimization variables. A diagonal matrix with a dimension equal to the number of energy storage units is constructed. Each element of this diagonal matrix is ​​taken as the response characteristic weight coefficient of the corresponding energy storage unit, where the weight coefficient of a fast-response unit is 0.1 and the weight coefficient of a slow-response unit is 0.9. This diagonal matrix is ​​the response characteristic weight matrix, which is multiplied into the overall lifetime of the energy storage system in the multi-objective function. In the loss minimization sub-objective, the power change of slow-response units is penalized more severely. The prediction time domain is set to ten control cycles, each control cycle being one second. The control time domain is set to be equal to the prediction time domain. The power value of each energy storage unit in each cycle is used as the power allocation value, and the initial power allocation value is taken as the actual output power of each energy storage unit at the current moment. A sequential quadratic programming algorithm is used for iterative solution. In each iteration, the industrial cloud platform calculates the multi-objective function value based on the current power allocation value and adjusts the power allocation value according to the constraint boundary to prevent it from going out of bounds. At the same time, the response characteristic weight matrix is ​​used to adjust the search direction. The iteration is terminated when the absolute value of the change rate of the multi-objective function value between two adjacent iterations is less than one-thousandth, or when the number of iterations exceeds fifty. The power allocation value obtained at the time of termination is organized into a sequence according to the energy storage unit and time order. This sequence is used as the optimal power control sequence for each energy storage unit in the control time domain.

[0045] In some embodiments, the process of sending instructions for the optimal power control sequence to the corresponding energy storage unit based on the industrial cloud platform can be achieved through the following steps: In the industrial cloud platform, virtual device objects are constructed that correspond one-to-one with each energy storage unit. The virtual device objects store the communication protocol, address mapping, and power response delay parameters of the corresponding energy storage unit. The optimal power control sequence is split into single-step instructions according to timestamps, and the instruction issuance time is pre-compensated according to the power response delay parameters of each energy storage unit. The pre-compensated instructions are sent to the local controller of the corresponding energy storage unit through the edge gateway of the industrial cloud platform, and the local controller drives the power conversion unit to execute them.

[0046] It should be noted that the virtual device object in this application is a digitally mapped entity created for each physical energy storage unit in the industrial cloud platform. This entity encapsulates all protocol information and timing parameters required for communication with the energy storage unit. The communication protocol is a set of specifications defining the data exchange format, transmission rules, and handshake process between the industrial cloud platform and the local controller of the energy storage unit. The address mapping is a conversion table used to associate the physical addresses of the internal registers of the energy storage unit with logical variables in the virtual device object. The power response delay parameter is a value used to quantify the time interval required for the energy storage unit to reach the command value from receiving the power command. The single-step command is a standardized data frame used to instruct the energy storage unit to output active power within a single control cycle. Pre-compensation is a data processing operation used to advance the actual command issuance time based on the power response delay parameter, so that the actual power response time of the energy storage unit is aligned with the timestamp specified in the optimal power control sequence. The local controller is an edge-side control device used to receive commands issued by the industrial cloud platform, perform power closed-loop regulation, and transmit the execution results back. A power conversion unit is a power electronic device used to convert the DC power of an energy storage unit into AC power or vice versa, based on the pulse width modulation signal output by the local controller.

[0047] In specific implementation, virtual device objects corresponding to each energy storage unit are constructed in the industrial cloud platform. These virtual device objects store the communication protocol, address mapping, and power response delay parameters of the corresponding energy storage unit. Specifically, the industrial cloud platform creates a corresponding virtual device object in the platform's device management module based on the factory identifier of each physical energy storage unit in the energy storage system. This virtual device object is assigned a unique internal identification code. For each virtual device object, the industrial cloud platform writes three key parameters from a pre-stored access configuration table: the first parameter is the communication protocol type, which is determined based on the bus standard supported by the local controller of the energy storage unit; the second parameter is the address mapping table, which records the starting address of the register used to receive power commands in the local controller, the command format conversion rules, and the address of the status feedback register; the third parameter is the power response delay parameter, which is obtained by performing a step response test on the energy storage unit, specifically the time elapsed from the moment the command is issued until the actual power of the energy storage unit reaches 90% of the command value, in milliseconds. All three parameters are stored in the corresponding virtual device objects, and all virtual device objects are used as calling interfaces for subsequent command splitting.

[0048] In specific implementation, the optimal power control sequence is split into single-step instructions according to timestamps, and the instruction issuance time is pre-compensated according to the power response delay parameters of each energy storage unit. Specifically, the industrial cloud platform extracts the power setpoints of all energy storage units in each control cycle from the optimal power control sequence according to the timestamp of each control cycle. For each energy storage unit, the corresponding timestamp and power setpoint are encapsulated into a single-step instruction according to the frame format specified by the communication protocol stored in the virtual device object of the energy storage unit. Then, the industrial cloud platform reads the power response delay parameters in the virtual device object of the energy storage unit, subtracts the delay value from the original issuance time of the single-step instruction, and obtains the pre-compensated issuance time. The above decapsulation and pre-compensation operations are repeated for each timestamp and each energy storage unit in the optimal power control sequence, and the pre-compensated issuance time is bound with the corresponding single-step instruction to form a queue of instructions to be issued.

[0049] In practice, the pre-compensated instructions are sent to the local controller of the corresponding energy storage unit through the edge gateway of the industrial cloud platform. The local controller then drives the power conversion unit to execute the instructions. Specifically, the industrial cloud platform transmits the single-step instructions in the instruction queue and the corresponding pre-compensated issuance time to the edge gateway device deployed on the field side of the energy storage system. The edge gateway device has a built-in real-time clock. When the system time reaches the pre-compensated issuance time of a certain single-step instruction, the edge gateway device reads the communication protocol and address mapping information in the virtual device object of the energy storage unit corresponding to the instruction, and writes the power setpoint to the local controller specified in the address mapping according to the protocol format. Register address; After receiving a new power setpoint, the local controller compares it with its own real-time measured actual output power, calculates the control signal using its internal proportional-integral regulation algorithm, and sends the control signal to the drive circuit of the power conversion unit in pulse-width modulation form; The insulated-gate bipolar transistor in the power conversion unit turns on or off according to the drive signal, converting the DC bus voltage of the energy storage unit into an AC voltage synchronized with the power grid, thereby controlling the actual output power of the energy storage unit to follow the command value; The local controller simultaneously transmits the actual power value, execution status, and current state of charge back to the industrial cloud platform through the edge gateway to complete closed-loop execution.

[0050] In some embodiments, reference Figure 3This figure is a flowchart of the closed-loop intelligent control process of the energy storage system in some embodiments of this application. As shown in the figure, the figure illustrates the collaborative control logic of multiple types of energy storage units applied to the industrial cloud platform: The industrial cloud platform, as the core carrier for data interaction and command issuance, first collects the real-time operation data and historical power data of the energy storage system. After the two types of data enter the data acquisition stage, they are input to the trajectory construction module. This module constructs a power reference trajectory based on the long-term trend of historical power data and the differentiated response characteristics of each energy storage unit. It also generates a dynamic feedback correction amount by combining the state of charge of each energy storage unit in the real-time operation data and the preset healthy operation center value, and performs closed-loop correction on the power reference trajectory. The corrected power trajectory is input to the power output module, which iteratively optimizes it through the power safety constraints of each energy storage unit to generate the optimal power control sequence for each energy storage unit. Finally, the generated power control sequence command is transmitted back to the industrial cloud platform through the data transmission link, and then issued by the platform to the corresponding energy storage unit for execution.

[0051] Furthermore, in another aspect of this application, in some embodiments, this application provides an intelligent control system for the operating status of an energy storage system, with reference to... Figure 4 The figure is a schematic diagram of the structure of an intelligent control system for the operation status of an energy storage system according to some embodiments of this application. The intelligent control system 400 for the operation status of the energy storage system includes: an acquisition module 401, a processing module 402, and an execution module 403, which are described below: The acquisition module 401 in this application is mainly used to acquire real-time operating data and historical power data of the energy storage system through the industrial cloud platform. Processing module 402, in this application, is used to construct a power reference trajectory that the energy storage system is expected to exchange at the grid connection point with the power grid based on the long-term power variation trend in historical power data and the response characteristics of each energy storage unit. It should be noted that the processing module 402 in this application is also used to generate dynamic feedback correction amounts for the state of charge of each energy storage unit and the preset central value characterizing the long-term healthy operation state of the energy storage system based on the state of charge of each energy storage unit in the real-time operation data, and to perform closed-loop correction on the power reference trajectory based on all the dynamic feedback correction amounts to obtain the corrected power reference trajectory. The execution module 403 in this application is mainly used to iteratively optimize the power of the energy storage system through the power safety constraints of each energy storage unit and the modified power reference trajectory, to obtain the optimal power control sequence of each energy storage unit in the energy storage system, and to issue the instruction of the optimal power control sequence to the corresponding energy storage unit for execution based on the industrial cloud platform.

[0052] In addition, this application also provides a computer device, which includes a memory and a processor. The memory stores code, and the processor is configured to acquire the code and execute the above-described intelligent control method for the operating state of the energy storage system.

[0053] In some embodiments, reference Figure 5 The figure is a schematic diagram of the structure of a computer device for implementing an intelligent control method for the operating state of an energy storage system according to some embodiments of this application. The intelligent control method for the operating state of the energy storage system in the above embodiments can... Figure 5 The computer device shown is used to implement this, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0054] Processor 501 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0055] The communication bus 502 can be used to transmit information between the aforementioned components.

[0056] Memory 503 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 503 may exist independently and be connected to processor 501 via communication bus 502. Memory 503 may also be integrated with processor 501.

[0057] The memory 503 stores program code for executing the scheme of this application, and its execution is controlled by the processor 501. The processor 501 executes the program code stored in the memory 503. The program code may include one or more software modules. The method used in the above embodiments can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0058] Communication interface 504 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0059] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0060] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0061] In addition, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described intelligent control method for the operating state of an energy storage system.

[0062] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0063] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. An intelligent control method for the operating status of an energy storage system, applied to an industrial cloud platform, wherein the energy storage system comprises at least two types of energy storage units with different response characteristics, characterized in that... The method includes the following steps: The industrial cloud platform is used to obtain real-time operating data and historical power data of the energy storage system. Based on the long-term power variation trend in historical power data and the response characteristics of each energy storage unit, a power reference trajectory is constructed for the expected power exchange between the energy storage system and the grid connection point. Based on the state of charge of each energy storage unit in the real-time operating data and the preset central value that characterizes the long-term healthy operating state of the energy storage system, a dynamic feedback correction amount for the state of each energy storage unit is generated. The power reference trajectory is then closed-loop corrected according to all the dynamic feedback correction amounts to obtain the corrected power reference trajectory. The power of the energy storage system is iteratively optimized by using the power safety constraints of each energy storage unit and the corrected power reference trajectory to obtain the optimal power control sequence of each energy storage unit in the energy storage system. The instructions of the optimal power control sequence are then sent to the corresponding energy storage units for execution based on the industrial cloud platform. Specifically, the optimal power control sequence for each energy storage unit in the energy storage system is obtained by iteratively optimizing the power of the energy storage system using the power safety constraints of each energy storage unit and the corrected power reference trajectory. A multi-objective function is established with the goal of minimizing tracking error, balancing the state of charge of each energy storage unit, and minimizing the overall lifespan loss of the energy storage system. The corrected power reference trajectory is used as the tracking target, and the upper and lower limits of power, the upper limit of power change rate, and the state of charge boundary of each energy storage unit are used as power safety constraints. Using the power safety constraints as the boundary, the multi-objective function is solved by rolling optimization, and the response characteristic weight matrix of each energy storage unit is introduced. The power allocation value in multiple time domains in the future is calculated iteratively. When the rate of change of the multi-objective function between two adjacent iterations is less than a preset threshold or the maximum number of iterations is reached, the iteration is terminated, and the optimal power control sequence of each energy storage unit in the control time domain is output. Specifically, the process of issuing instructions for the optimal power control sequence to the corresponding energy storage unit based on the industrial cloud platform includes: In the industrial cloud platform, virtual device objects are constructed that correspond one-to-one with each energy storage unit. The virtual device objects store the communication protocol, address mapping, and power response delay parameters of the corresponding energy storage unit. The optimal power control sequence is split into single-step instructions according to timestamps, and the instruction issuance time is pre-compensated according to the power response delay parameters of each energy storage unit. The pre-compensated instructions are sent to the local controller of the corresponding energy storage unit through the edge gateway of the industrial cloud platform, and the local controller drives the power conversion unit to execute them.

2. The method as described in claim 1, characterized in that, The real-time operating data includes the voltage, current, state of charge, temperature, and health status of each energy storage unit, and the historical power data includes the day-ahead, intraday, and weekly power time-series data of the energy storage system at the grid connection point.

3. The method as described in claim 1, characterized in that, Based on the long-term power variation trend in historical power data and the response characteristics of each energy storage unit, a power reference trajectory for the expected exchange between the energy storage system and the grid connection point is constructed, specifically including: Empirical mode decomposition is performed on the historical power data to extract its long-term trend components and short-term fluctuation components; Obtain the response characteristics of each energy storage unit; Based on the response characteristics of each energy storage unit, the long-term trend component is allocated to the energy storage unit with a slower response speed, and the short-term fluctuation component is allocated to the energy storage unit with a faster response speed. The weighted summation of each allocated component is performed, and a power reference trajectory for the expected exchange between the energy storage system and the grid connection point is constructed based on the grid dispatch instructions and the voltage deviation constraint at the grid connection point.

4. The method as described in claim 1, characterized in that, The dynamic feedback correction amount for the state of each energy storage unit, generated based on the state of charge of each energy storage unit in the real-time operating data and the preset central value characterizing the long-term healthy operating state of the energy storage system, specifically includes: Obtain the safe operating range of the state of charge of each energy storage unit and the preset central value; The deviation between the state of charge of each energy storage unit and the preset central value is calculated to obtain the basic correction component of each energy storage unit; Based on the differences in response characteristics of each energy storage unit and the safe operating range of the state of charge, the basic correction components of each energy storage unit are matched in time scale and limited in amplitude to generate dynamic feedback correction quantities for the state of each energy storage unit.

5. The method as described in claim 1, characterized in that, Based on all dynamic feedback corrections, a closed-loop correction is performed on the power reference trajectory to obtain the corrected power reference trajectory, which specifically includes: The total correction amount is obtained by vector summation of the dynamic feedback corrections of all energy storage units after time synchronization. The total correction amount is superimposed on the power reference trajectory to form a preliminary correction trajectory; By using the power change rate constraint at the grid connection point to limit the slope of the preliminary correction trajectory, a corrected power reference trajectory that meets the grid safety requirements is obtained.

6. An intelligent control system for the operating status of an energy storage system, which employs the method described in any one of claims 1 to 5 for intelligent control, and is applied to an industrial cloud platform, wherein the energy storage system comprises at least two types of energy storage units with different response characteristics, characterized in that, The system includes: The acquisition module is used to acquire real-time operating data and historical power data of the energy storage system through the industrial cloud platform; The processing module is used to construct the power reference trajectory that the energy storage system is expected to exchange at the grid connection point with the grid based on the long-term power variation trend in historical power data and the response characteristics of each energy storage unit. The processing module is also used to generate dynamic feedback correction quantities for the state of charge of each energy storage unit and the preset central value characterizing the long-term healthy operation state of the energy storage system based on the state of charge of each energy storage unit in the real-time operation data, and to perform closed-loop correction on the power reference trajectory based on all the dynamic feedback correction quantities to obtain the corrected power reference trajectory. The execution module is used to iteratively optimize the power of the energy storage system by using the power safety constraints of each energy storage unit and the corrected power reference trajectory to obtain the optimal power control sequence of each energy storage unit in the energy storage system, and to issue the instructions of the optimal power control sequence to the corresponding energy storage unit for execution based on the industrial cloud platform.

7. A computer device, characterized in that, The computer device includes a memory and a processor, the memory storing code, and the processor being configured to retrieve the code and execute the intelligent control method for the operating state of the energy storage system as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the intelligent control method for the operating state of the energy storage system as described in any one of claims 1 to 5.

Images