Energy storage system power distribution method and system based on proportional integral control

Abstract

The application relates to the technical field of energy storage system control, and discloses an energy storage system power distribution method and system based on proportional integral control, which comprises collecting the real-time state of charge and the maximum power of each battery cluster, judging whether the battery cluster is in a charging mode or a discharging mode, and setting a target state of charge; calculating a state deviation, using a proportional integral controller to calculate an original power correction amount and subtracting the average value to obtain a zero-normalized power correction amount; calculating a reference power corresponding to a scheduling instruction, superimposing the zero-normalized power correction amount to obtain a preliminary power instruction, determining the limit when the limit power is proportionally distributed to the non-limit battery cluster to obtain a final power instruction; and issuing the instruction for execution and freezing the integral operation when the power boundary is touched. The application ensures that the power sum is equal to the scheduling requirement through zero-normalization processing, guarantees the safe operation of hardware through the distribution of the limit power, and eliminates the desaturation delay through the freezing of the integral operation.

Description

Technical Field

[0001] This invention relates to the field of energy storage system control technology, specifically to a power distribution method and system for energy storage systems based on proportional-integral control. Background Technology

[0002] During the operation of a string energy storage system, the state of charge (SOC) of multiple independent battery clusters is unbalanced. When using a proportional-integral (PI) controller to calculate the power correction for each battery cluster for internal SOC balancing, the sum of the power corrections generated by each control loop is not zero. After adding the unprocessed power correction to the base power, the total power allocated to each battery cluster deviates from the overall SOC power dispatch command issued by the external power grid.

[0003] In actual operation, to maintain a constant total charging and discharging power to strictly respond to external grid dispatch, when some fully charged or discharged battery clusters prematurely exit due to imbalanced state of charge, the power borne by the remaining online battery clusters will passively surge. This power surge can easily trigger the maximum power limit of a single remaining battery cluster. To avoid exceeding the limits of the underlying hardware, the existing conventional hard numerical truncation process will cause the system to operate at the maximum power that the remaining clusters can carry, and the power originally allocated to the over-limited battery clusters will be directly discarded. This results in the total output power of the energy storage system failing to strictly follow the overall dispatch demand of the grid, thus severely reducing the overall charging and discharging capacity of the energy storage system and the arbitrage benefits of the project.

[0004] Therefore, this invention proposes a power allocation method and system for energy storage systems based on proportional-integral control to address the shortcomings of existing technologies. Summary of the Invention

[0005] The technical problem solved by this invention is that the state of charge of each battery cluster in a string energy storage system is inconsistent during charging and discharging. The power allocation strategy deviates from the total power scheduling target when balancing the state of charge of each battery cluster. At the same time, due to the inconsistent state of charge, some fully charged or discharged battery clusters will be disconnected in advance, which forces the power undertaken by the remaining online battery clusters to passively increase, thereby triggering the maximum power limit of a single cluster. As a result, the actual total charging and discharging power of the system cannot strictly follow the scheduling command, which ultimately leads to a reduction in the overall charging and discharging capacity of the energy storage system and a reduction in arbitrage profits.

[0006] To address the problems existing in the prior art, the present invention provides the following technical solution: The first aspect of this invention provides a power allocation method for an energy storage system based on proportional-integral control, comprising the following steps: Collect the real-time state of charge and maximum power of each battery cluster in the energy storage system to determine whether the energy storage system is in charging or discharging mode, and set the target state of charge based on the determined mode. Calculate the state deviation between the real-time state of charge and the target state of charge for each battery cluster. Use a proportional-integral controller to calculate the original power correction based on the state deviation. Calculate the average of the original power corrections for all battery clusters. Subtract the average from the original power correction to obtain the zeroed power correction. Receive the total power scheduling command for the entire cabin, calculate the reference power corresponding to the total power scheduling command for the entire cabin, add the reference power and the power correction amount after zeroing to obtain the preliminary power command, check whether the preliminary power command is greater than the maximum power or less than 0, if it is determined that there is an over-limit situation, calculate the over-limit power and allocate the over-limit power proportionally to the non-over-limit battery clusters to obtain the final power command; The final power command is sent to the energy storage converter corresponding to each battery cluster for execution, and the integral accumulation calculation of the corresponding battery cluster is frozen when the final power command reaches the power boundary.

[0007] Determine whether the energy storage system is in charging or discharging mode, and set the target state of charge based on the determined mode, including: When the energy storage system is determined to be in charging mode, it iterates through the real-time state of charge data of all received battery clusters, filters out the real-time state of charge with the largest value, and sets the real-time state of charge with the largest value as the target state of charge. When the energy storage system is determined to be in discharge mode, the system iterates through the real-time state of charge data of all received battery clusters, selects the real-time state of charge with the smallest value, and sets the real-time state of charge with the smallest value as the target state of charge.

[0008] Subtracting the mean from the original power correction amount yields the zeroed power correction amount, which in turn causes the sum of the zeroed power correction amounts of all battery clusters to equal 0, ensuring that the total power allocated to each battery cluster is equal to the total power scheduling command for the entire cabin.

[0009] The process of calculating the over-limit power and proportionally allocating it to the non-over-limit battery clusters yields the final power command, including: Extract the preliminary power command of each battery cluster, compare the preliminary power command with the maximum power of the corresponding battery cluster, and simultaneously compare the preliminary power command with 0. When the initial power command is greater than the maximum power, it is determined that the corresponding battery cluster has exceeded the upper limit. The initial power command is subtracted from the maximum power to obtain the over-limit power of the corresponding battery cluster, and the maximum power is set as the temporary command value after the corresponding battery cluster is truncated. When the initial power command is less than 0, it is determined that the corresponding battery cluster has exceeded the lower limit. The initial power command is subtracted from 0 to obtain the over-limit power of the corresponding battery cluster, and 0 is set as the temporary command value after the corresponding battery cluster is truncated. When the initial power command is less than or equal to the maximum power and greater than or equal to 0, it is determined that the corresponding battery cluster is in the safe operating range and no over-limit situation has occurred. The over-limit power of the corresponding battery cluster is set to 0, and the initial power command is directly set to the temporary command value after the corresponding battery cluster is truncated. The total over-limit power is obtained by summing the over-limit power of all battery clusters that have exceeded the limit; the temporary instruction values ​​corresponding to the battery clusters that are in the safe operating range and have not exceeded the limit are extracted, and the temporary instruction values ​​corresponding to all battery clusters that have not exceeded the limit are summed to obtain the total non-over-limit instruction value. Calculate the allocation ratio corresponding to each battery cluster that has not exceeded the limit. The allocation ratio is equal to the temporary instruction value of the corresponding battery cluster that has not exceeded the limit divided by the sum of the non-limited instructions. Multiply the total over-limit power by the allocation ratio to obtain the additional redistribution power received by the battery cluster that has not exceeded the limit. Add the temporary command value and the additional redistribution power to obtain the new round of power command corresponding to the battery cluster that has not exceeded the limit. The new round of power commands is used as the initial power command input to the limit detection stage for loop boundary judgment. The loop stops when all commands of all battery clusters are between the maximum power and 0. The power commands of all battery clusters that are in the safe operating range when the loop stops are uniformly set as the final power commands.

[0010] When the final power command reaches the power boundary, the integral accumulation operation of the corresponding battery cluster is frozen, including: when it is determined that the final power command is equal to the maximum power or equal to 0, a saturation intervention signal is sent to the proportional-integral controller of the corresponding battery cluster to forcibly stop the state deviation of the corresponding battery cluster in the current sampling period from entering the integration stage for accumulation operation, and the state deviation input to the proportional-integral controller is forcibly set to zero at the same time.

[0011] A second aspect of the present invention provides a power distribution system for an energy storage system based on proportional-integral control, comprising: The status setting module is used to collect the real-time state of charge of each battery cluster in the energy storage system and set the target state of charge. The correction calculation module is used to calculate the state deviation between the real-time state of charge of each battery cluster and the target state of charge, and to calculate the zeroed power correction amount based on the state deviation. The over-limit allocation module is used to receive the total power scheduling command for the entire cabin, calculate the reference power corresponding to the total power scheduling command for the entire cabin, add the reference power and the zeroed power correction amount to obtain the preliminary power command, and when it is determined that the preliminary power command has an over-limit situation, allocate the over-limit power to obtain the final power command. The instruction execution module is used to send the final power instruction to the energy storage converter corresponding to each battery cluster for execution and freeze the integral accumulation calculation of the corresponding battery cluster when the final power instruction reaches the power boundary.

[0012] This invention provides a power allocation method and system for energy storage systems based on proportional-integral control. It offers the following advantages: 1. This invention calculates the average of the original power correction values ​​of all battery clusters and subtracts the average value from the original power correction values ​​to obtain the zeroed power correction values. This ensures that the sum of the zeroed power correction values ​​of all battery clusters equals 0. This ensures that while dynamically adjusting the power of each cluster during charging and discharging and promoting the uniformity of the state of charge of each cluster, the total power allocated to each battery cluster is equal to the total power dispatch command of the entire module. This ensures that the system does not deviate from the overall dispatch requirements issued by the external power grid during the process of adjusting the state of charge of the battery clusters.

[0013] 2. This invention detects whether the initial power command is greater than the maximum power or less than 0. When an over-limit situation is determined, the total over-limit power is calculated. Based on the allocation ratio corresponding to each non-over-limit battery cluster, the total over-limit power is allocated to the non-over-limit battery clusters to obtain a new round of power commands. The cycle boundary judgment is performed until all battery cluster commands are between the maximum power and 0. When a local node reaches the operating limit, the remaining power demand is transferred to the battery clusters within the safe operating range. This solves the engineering pain point of passively increasing the power of the remaining online clusters and triggering the power limitation of a single cluster due to the early exit of some fully charged or discharged clusters. It avoids the overall power derating of the energy storage system, ensures the accurate execution of the grid's overall dispatch command, thereby maximizing the actual charging and discharging capacity of the system and improving the economic arbitrage benefits of the project.

[0014] 3. This invention sends a saturation intervention signal to the proportional-integral controller of the corresponding battery cluster when the final power command is equal to the maximum power or equal to 0. This forces the state deviation to stop entering the integration stage for accumulation calculation and simultaneously forces the state deviation to zero. This cuts off the saturation node command drift path caused by the battery cluster that has not exceeded the limit, maintains the original power correction amount constant and avoids deep integration saturation. It eliminates the desaturation delay caused by integration saturation and ensures the command tracking speed of the energy storage system under the condition of frequent power boundary switching. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the architecture of the present invention; Figure 2 This is a flowchart of the method of the present invention; Figure 3 This is a schematic diagram of a sub-process of step S10 of the present invention; Figure 4 This is a schematic diagram of a sub-process of step S20 of the present invention; Figure 5 This is a schematic diagram of a sub-process of step S30 of the present invention; Figure 6This is a schematic diagram of a sub-process of step S40 of the present invention; Figure 7 This is a comparison diagram of the real-time state of charge convergence of the battery clusters according to the present invention. Detailed Implementation

[0016] The technical solutions in 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.

[0017] See attached document Figure 1 The present invention provides a string energy storage system, comprising: An energy management system, N battery clusters, N energy storage converters, and a power grid.

[0018] Each battery cluster is connected to its corresponding energy storage converter to form an independent energy storage unit. The AC side of each energy storage converter is connected in parallel to the AC bus, and the AC bus is connected to the power grid via a step-up transformer.

[0019] The energy management system sends the total power dispatch command to the central controller. The central controller generates a power allocation command based on the real-time state of charge data and power limits uploaded by each battery cluster and sends it to the corresponding energy storage converter of each battery cluster for execution.

[0020] See attached document Figure 2 This invention provides a power allocation method for an energy storage system based on proportional-integral control, comprising the following steps: S10: Collect the real-time state of charge and maximum power of each battery cluster in the energy storage system, determine whether the energy storage system is in charging mode or discharging mode, and set the target state of charge according to the determined mode. S20, calculate the state deviation between the real-time state of charge of each battery cluster and the target state of charge, use the proportional-integral controller to calculate the original power correction based on the state deviation, calculate the average of the original power correction of all battery clusters, and subtract the average from the original power correction to obtain the zeroed power correction. S30: Receive the total power scheduling command for the entire cabin, calculate the reference power corresponding to the total power scheduling command for the entire cabin, add the reference power and the zeroed power correction amount to obtain the preliminary power command, detect whether the preliminary power command is greater than the maximum power or less than 0, if it is determined that there is an over-limit situation, calculate the over-limit power and allocate the over-limit power proportionally to the non-over-limit battery clusters to obtain the final power command. S40 sends the final power command to the energy storage converter corresponding to each battery cluster for execution and freezes the integral accumulation calculation of the corresponding battery cluster when the final power command reaches the power boundary.

[0021] This invention also provides a power distribution system for energy storage systems based on proportional-integral control, comprising: The status setting module is used to collect the real-time state of charge of each battery cluster in the energy storage system and set the target state of charge. The correction calculation module is used to calculate the state deviation between the real-time state of charge of each battery cluster and the target state of charge, and to calculate the zeroed power correction amount based on the state deviation. The over-limit allocation module is used to receive the total power scheduling instruction for the entire cabin, calculate the reference power corresponding to the total power scheduling instruction for the entire cabin, add the reference power and the zeroed power correction amount to obtain the preliminary power instruction, and when it is determined that the preliminary power instruction has an over-limit situation, allocate the over-limit power to obtain the final power instruction. The instruction execution module is used to send the final power instruction to the energy storage converter corresponding to each battery cluster for execution.

[0022] The following section elaborates on each step of the power allocation method for string energy storage systems based on proportional-integral control.

[0023] See attached document Figure 3 The S10 step includes the following sub-steps: S101, the central controller receives real-time state of charge (SOC) data reported by the battery management systems of each battery cluster via a communication interface. The energy storage system has multiple independently operating battery clusters, and the battery management system is installed inside each battery cluster to monitor its status. The real-time SOC data reflects the current remaining charge percentage of the battery cluster, and the variable is expressed using the following formula.

[0024] ; In the above formula, Representing the Each battery cluster during the sampling period Real-time state of charge; Represents the battery cluster serial number. The value ranges from 1 to Positive integers; Represents the total number of battery clusters, and ; This represents the current sampling period of the discrete operation of the control system.

[0025] The specific process by which a battery management system collects the voltage and current of individual battery cells and calculates the real-time state of charge is implemented by those skilled in the art using the ampere-hour integration method or the Kalman filter algorithm. The calculation process is a well-known technology in the field and will not be described in detail here.

[0026] S102, the central controller synchronously receives the maximum power data sent by the battery management systems of each battery cluster. The battery management system dynamically evaluates the charge and discharge capabilities based on the current temperature and terminal voltage of the battery cells to obtain the maximum power data and sends it to the central controller. The maximum power data represents the safe power limit that the battery cluster can withstand under the current environment and current charge level, and the variable is expressed by the following formula.

[0027] ; In the above formula, Representing the Each battery cluster during the sampling period The maximum power that the battery management system allows to operate at.

[0028] S103, the central controller receives the total power dispatch command for the entire cabin from the energy management system via the communication network. The total power dispatch command for the entire cabin includes a positive or negative sign parameter. The positive or negative sign parameter indicates the desired direction of energy flow.

[0029] For the process of parsing communication message data, those skilled in the art can write underlying communication protocol decomposition code to extract symbol bit variables. Communication message parsing is a well-known technology in this field and will not be elaborated here.

[0030] S104, the central controller extracts and determines the sign of the parameter within the total power dispatch command for the entire storage module. When the sign is positive, the energy flow direction is from the grid side into the energy storage system, and the central controller determines that the energy storage system is in charging mode. When the sign is negative, the energy flow direction is from the energy storage system out to the grid side, and the central controller determines that the energy storage system is in discharging mode. When the value of the total power dispatch command for the entire storage module is equal to 0, the central controller determines that the energy storage system is in standby mode.

[0031] In addition, when the central controller detects a change in the operating mode of the energy storage system (including switching from charging mode to discharging mode, switching from discharging mode to charging mode, or re-entering charging / discharging mode after standby mode), the central controller triggers the integral reset mechanism, sends an integral clear signal to the proportional-integral controllers corresponding to all battery clusters, and clears the integral accumulation register inside the proportional-integral controller to zero.

[0032] S105, when the energy storage system is determined to be in charging mode, the central controller iterates through the real-time state of charge (SOC) data of all received battery clusters and selects the SOC with the highest value. The central controller sets the selected SOC as the target SOC. In charging mode, the SOC with the highest value is selected as the dynamic catch-up benchmark, causing battery clusters with lower charge levels to accelerate charging to catch up with the battery clusters with higher charge levels, thus ensuring that all battery clusters reach the full charge boundary synchronously. The process of setting the target SOC in charging mode uses the following formula.

[0033] ; In the above formula, Represents the sampling period The target state of charge is determined at that time; This represents the function that takes the maximum value. Representing the Each battery cluster during the sampling period Real-time state of charge.

[0034] S106, when the energy storage system is determined to be in discharge mode, the central controller iterates through the real-time state of charge (SOC) data of all received battery clusters and selects the SOC with the smallest value. The central controller sets the selected SOC with the smallest value as the target SOC. In discharge mode, the SOC with the smallest value is selected as the dynamic catch-up benchmark, causing the battery clusters with higher charge to accelerate their discharge to catch up with the battery clusters with lower charge, thus prompting all battery clusters to reach the discharge boundary synchronously. The process of setting the target SOC in discharge mode uses the following formula.

[0035] ; In the above formula, Represents the function that takes the minimum value; Represents the sampling period The target state of charge is determined at that time; Representing the Each battery cluster during the sampling period Real-time state of charge.

[0036] See attached document Figure 4 The S20 step includes the following sub-steps: S201, when the energy storage system is in charging mode, the central controller subtracts the real-time state of charge of the nth battery cluster from the target state of charge to obtain the state deviation of the corresponding battery cluster. The difference calculated in charging mode represents the distance between the current battery cluster and the highest capacity reference. The calculation process of the state deviation in charging mode uses the following formula.

[0037] ; In the formula, Representing the Each battery cluster during the sampling period State deviation at that time; Represents the sampling period The target state of charge is determined at that time; Representing the Each battery cluster during the sampling period Real-time state of charge.

[0038] S202, when the energy storage system is in discharge mode, the central controller subtracts the target state of charge from the real-time state of charge of the nth battery cluster to obtain the state deviation of the corresponding battery cluster. The difference calculated in discharge mode represents the distance between the current battery cluster and the minimum charge benchmark. The calculation process of the state deviation in discharge mode uses the following formula.

[0039] ; In the formula, Representing the Each battery cluster during the sampling period State deviation at that time; Representing the Each battery cluster during the sampling period Real-time state of charge; Represents the sampling period The target state of charge is determined at that time.

[0040] S203, the central controller inputs the state deviation corresponding to each battery cluster into the corresponding proportional-integral controller for mathematical calculation. The proportional-integral controller receives the state deviation as input, uses the proportional element to multiply the state deviation by the proportional parameter to generate the first power adjustment component to respond to the current deviation state, and uses the integral element to perform time-dimensional cumulative calculation of the state deviation and multiply it by the integral parameter to generate the second power adjustment component to eliminate the steady-state error of the system.

[0041] S204, the proportional-integral controller adds the first power regulation component output from the proportional element and the second power regulation component output from the integral element to obtain the second power regulation component. The original power correction amount corresponding to each battery cluster. The process of calculating the original power correction amount using a proportional-integral controller is as follows.

[0042] ; In the formula, Representing the Each battery cluster during the sampling period The original power correction amount at that time; The proportional parameter representing the proportional-integral controller; Representing the Each battery cluster during the sampling period State deviation at that time; The integral parameters represent the proportional-integral controller. This represents the integral of the state deviation over the time dimension.

[0043] When a control system performs discretization operations in a digital processing chip, the process of transforming the integral element into an accumulation and summation can be implemented by writing software programs using the backward Euler method or the forward Euler method. The conversion of continuous domain control algorithms to discrete domain is a well-known technology in this field and will not be elaborated here.

[0044] S205: After obtaining the original power correction values ​​corresponding to all battery clusters, the central controller sums all the original power correction values ​​to obtain the total correction value. The average value of the original power correction value is obtained by dividing the total correction value by the total number of battery clusters. The process of calculating the average value uses the following formula.

[0045] ; In the formula, Represents the sampling period The mean of the original power correction for all battery clusters at that time; Represents the total number of battery clusters; Represents the battery cluster serial number; Representing the Each battery cluster during the sampling period The original power correction amount at that time.

[0046] S206, the central controller will... Subtracting the calculated mean from the original power correction value for each battery cluster yields the zeroed power correction value for that battery cluster. The process of calculating the zeroed power correction value uses the following formula.

[0047] ; In the formula, Representing the Each battery cluster during the sampling period Power correction amount after time-reset; Representing the Each battery cluster during the sampling period The original power correction amount at that time; Represents the sampling period The mean of the original power correction for all battery clusters at that time.

[0048] In step S207, the central controller performs a subtraction operation to reduce the mean, ensuring that the sum of the zeroed-out power corrections for all battery clusters equals zero. In subsequent stages, when the zeroed-out power corrections are added to the base power, this zero-sum characteristic guarantees that the final power allocated to each battery cluster is exactly equal to the overall power dispatch command for the entire module. The central controller, through mean elimination, ensures that the system's internal adjustments to the battery cluster state of charge do not deviate from the overall dispatch requirements issued by the external power grid.

[0049] See attached document Figure 5 The S30 step includes the following sub-steps: S301, the central controller receives the determined total power scheduling command for the entire battery cell from the energy management system. It extracts the absolute value of the total power scheduling command as the initial demand for subsequent power allocation calculations, unifying the numerical processing logic between charging and discharging modes. It counts the number of each battery cluster. The central controller sums the maximum power of all battery clusters to obtain the current maximum available power of the system. The central controller compares the initial demand with the current maximum available power of the system.

[0050] When the initial demand exceeds the system's current maximum available power, the system's current maximum available power will be set to the absolute value of the corresponding total cabin power scheduling command. When the initial demand is less than or equal to the current total available maximum power of the system, the initial demand will be set to the absolute value of the total power scheduling command for the entire cabin. The central controller will assign the absolute value of the total power dispatch command for the entire cabin. Dividing the total power demand by the total number of battery clusters, the power requirement is evenly distributed among all battery clusters within the system to obtain the baseline power allocated to each individual battery cluster. The baseline power is calculated using the following formula.

[0051] ; In the formula, Represents the sampling period The absolute value of the reference power obtained by calculation at that time; Represents the sampling period The absolute value of the total cabin power scheduling command after pre-limiting verification; This represents the total number of battery clusters.

[0052] In step S302, the central controller adds the evenly distributed reference power to the zeroed-out power correction amount corresponding to the nth battery cluster. The reference power plus the zeroed-out power correction amount yields the initial power command for the corresponding battery cluster. The initial power command includes a reference component to satisfy external scheduling and a correction component to balance the power differences between internal battery clusters. The calculation of the initial power command uses the following formula.

[0053] ; In the formula, Representing the Each battery cluster during the sampling period The initial power command synthesized in time; Represents the sampling period The reference power obtained by calculation at that time; Representing the Each battery cluster during the sampling period The power correction amount after time-reset to zero.

[0054] S303: The central controller extracts the preliminary power command from each battery cluster, compares the preliminary power command with the maximum power of the corresponding battery cluster, and simultaneously compares the preliminary power command with 0. The central controller then determines whether the preliminary power command is within the safe operating range. The upper boundary of the safe operating range is the maximum power, and the lower boundary is 0.

[0055] S304, when the initial power command exceeds the maximum power, the central controller determines that the corresponding battery cluster has exceeded its upper limit. The central controller subtracts the maximum power from the initial power command to obtain the over-limit power of the corresponding battery cluster, and sets the maximum power to the temporary command value after the corresponding battery cluster is truncated.

[0056] S305, when the initial power command is less than 0, the central controller determines that the corresponding battery cluster has exceeded the lower limit. The central controller subtracts 0 from the initial power command to obtain the over-limit power of the corresponding battery cluster, and sets 0 as the temporary command value after the corresponding battery cluster is truncated.

[0057] S306, when the initial power command is less than or equal to the maximum power and greater than or equal to 0, the central controller determines that the corresponding battery cluster is within the safe operating range and no limit violation has occurred. The central controller sets the limit violation power of the corresponding battery cluster to 0 and directly sets the initial power command to the temporary command value after the corresponding battery cluster has been truncated.

[0058] S307: The central controller traverses all battery clusters, extracts the over-limit power corresponding to each battery cluster that experienced upper and lower boundary over-limit conditions, and sums all the over-limit power to obtain the total over-limit power. The process of calculating the total over-limit power uses the following formula.

[0059] ; In the formula, Represents the sampling period The total over-limit power obtained by summing up the data at the same time; Represents the total number of battery clusters; Represents the battery cluster serial number; Representing the Each battery cluster during the sampling period The over-limit power calculated at that time.

[0060] S308: The central controller filters out battery clusters that are within the safe operating range and have not exceeded limits, and counts the number of battery clusters that have not exceeded limits. The central controller extracts the temporary instruction values ​​corresponding to the battery clusters that have not exceeded limits, and sums up the temporary instruction values ​​corresponding to all battery clusters that have not exceeded limits to obtain the total number of non-limit-exceeding instructions.

[0061] S309, the central controller checks whether the sum of the non-over-limit instructions Psum,temp(t) is equal to 0. If it is equal to 0, it further checks whether the number of non-over-limit battery clusters is greater than 0; if the number of non-over-limit battery clusters is greater than 0, the central controller divides the total over-limit power by the number of non-over-limit battery clusters to obtain the additional redistribution power; if the number of non-over-limit battery clusters is equal to 0, the additional redistribution power is directly set to 0; if the sum of the non-over-limit instructions is greater than 0, the following proportional redistribution process begins.

[0062] The central controller calculates the allocation ratio for each battery cluster that has not exceeded its limits. The allocation ratio equals the temporary command value for the corresponding battery cluster that has not exceeded its limits divided by the sum of the non-limit-exceeding commands. The central controller multiplies the total limit-exceeding power by the allocation ratio to obtain the additional redistributed power received by the battery clusters that have not exceeded their limits. It then adds the temporary command value and the additional redistributed power to obtain the new round of power commands for the battery clusters that have not exceeded their limits. The process of allocating the total limit-exceeding power according to the command proportion uses the following formula.

[0063] ; In the formula, Representing the Each battery cluster that did not exceed the limit was in the sampling period The absolute value of the new round of power commands synthesized in time; Representing the Each battery cluster that did not exceed the limit was in the sampling period The temporary instruction value after truncation processing; Represents the sampling period The total over-limit power obtained by summing up the data at the same time; Represents all battery clusters that have not exceeded the limit during the sampling period. The summation result of the temporary instruction values ​​at that time, and .

[0064] In S310, the central controller re-inputs the new round of power commands as the initial power command into the over-limit detection stage for cycle boundary judgment. When the central controller determines that the new round of power commands triggers a new boundary over-limit situation, it repeats the steps of calculating the over-limit power and redistributing it proportionally. When the central controller determines that all battery cluster commands are between the maximum power and 0, it stops the cycle. The central controller then sets the power commands corresponding to all battery clusters that are within the safe operating range at the time of stopping the cycle as the final power command.

[0065] For the logic control process of implementing numerical limit loop detection by writing software code inside the central controller, those skilled in the art can write low-level loop judgment condition statements to implement it. The loop judgment logic is a well-known technology in this field and will not be described in detail here.

[0066] See attached document Figure 6 The S40 step includes the following sub-steps: In step S401, after obtaining the final power command for each battery cluster, the central controller retrieves the positive and negative sign parameters of the total power scheduling command for the entire battery pack extracted in step S104. It combines these parameters with the final power command (absolute value) to generate a final execution command with a precise charging / discharging direction. Subsequently, it converts all the data from the final execution commands into digital signals conforming to industrial communication standards. The central controller then packages and encapsulates the converted digital signals into corresponding communication control messages according to the format requirements of the underlying communication protocol.

[0067] For the format encapsulation and packaging process of communication control messages, those skilled in the art can use the CAN bus protocol or Modbus protocol to write the underlying communication code to implement it. The packaging of communication data is a well-known technology in this field and will not be described in detail here.

[0068] In S402, the central controller sends encapsulated communication control messages to the individual energy storage converters corresponding to each battery cluster in the energy storage system via the internal control LAN bus. The underlying control unit of the energy storage converter receives the communication control messages of the corresponding node and parses the specific value of the final power command from the message data.

[0069] S403, the energy storage converter calculates the tracking error based on the final power command obtained from the analysis and the actual output power of the battery cluster. It then adjusts the duty cycle of the drive signals of its internal power semiconductor switching devices according to this tracking error. By changing the duty cycle of the pulse width modulation signal, the energy storage converter controls the power conversion process of the AC-DC converter circuit, ensuring that the actual output power of the energy storage converter tracks and matches the final power command. The energy storage converter then executes the power output response, completing the charging and discharging control of the corresponding battery cluster.

[0070] In S404, the central controller reads the final power command corresponding to each battery cluster in real time during each sampling cycle and performs a value equality check with the current maximum power of the corresponding battery cluster and 0. Through this check, the central controller detects whether the energy storage converter corresponding to each battery cluster has reached its output limit, that is, whether the underlying actuator is in the power saturation range.

[0071] S405: When the central controller determines that the final power command of the nth battery cluster is equal to the maximum power or equal to 0, it determines that the corresponding underlying actuator is in a saturated state. The central controller triggers the anti-saturation mechanism and sends a saturation intervention signal to the proportional-integral controller of the corresponding battery cluster. After receiving the saturation intervention signal, the proportional-integral controller forcibly stops the state deviation of the corresponding battery cluster in the current sampling period from entering the integration stage for accumulation calculation, and simultaneously forces the state deviation input to the proportional-integral controller to zero.

[0072] S406, by forcibly setting the input source state deviation to zero under saturation, prevents the proportional-integral controller from generating new proportional adjustment increments and integral accumulation increments for the current saturation node. This algebraic blocking operation cuts off the saturation node command drift path caused by the superposition of global mean coupling between non-limited battery clusters within the system, maintaining the original power correction constant and avoiding deep integral saturation. The saturation intervention control process uses the following formula.

[0073] ; ; In the formula, Representing the The proportional-integral controller of each battery cluster in the current sampling period The value of the integral term at that time; Representing the The proportional-integral controller of each battery cluster in the previous sampling period The value of the integral term at that time; This represents the state deviation after being forcibly reset to zero. Both formulas hold true only if the final power command for the corresponding battery cluster is equal to or equal to the maximum power.

[0074] In step S407, when the central controller determines that the final power command is within the safe range of greater than 0 and less than the maximum power, the integral freeze signal is revoked, and the proportional-integral controller resumes normal integral accumulation. By implementing this integral freeze control, when the total system power dispatch command changes in the opposite direction, requiring the corresponding battery cluster to exit saturation, the proportional-integral controller can immediately respond to the new state deviation and calculate the effective power regulation component. This process eliminates the long desaturation delay caused by deep integral saturation in conventional control, ensuring the command tracking speed of the energy storage system under frequent power boundary switching conditions.

[0075] Specific application examples: The energy storage system is configured to include four battery clusters. The maximum power data for each battery cluster is The central controller received the total power dispatch command for the entire cabin. (Positive sign indicates charging mode). Current sampling period The real-time state of charge reported by each battery cluster is as follows: , , , .

[0076] Determined to be in charging mode, the target state of charge is set according to the formula: ; Calculate state deviation : ; Assuming the proportional parameter of the proportional-integral controller Ignoring the integral accumulation at the current instant, calculate the original power correction: ; Calculate the mean of the original power correction: ; Calculate the power correction amount after zeroing. : ; Calculate the reference power: ; The initial power command is obtained by adding the reference power and the zeroed power correction. : ; All initial power commands were greater than 0 and less than the maximum power (100kW), with no limit violations, and were uniformly set as the final power commands. Battery cluster 1, with the lowest battery level, received the maximum charging power, while battery cluster 4, with the highest battery level, received the minimum charging power. The total power... It perfectly tracks the total power scheduling instructions for the entire cabin.

[0077] Experimental verification and effect comparison: To verify the effectiveness of the present invention, a continuous charge-discharge test model of the energy storage system throughout its entire life cycle was built based on the above parameter boundaries, and the results were compared with those of the traditional average power allocation method without state deviation correction.

[0078] The conclusions are as follows: Reference Appendix Figure 7 The curves in the figure show that, under continuous charge-discharge scheduling, the real-time state of charge of the four battery clusters, which initially exhibited significant differences in charge levels, converged rapidly to the same baseline over time after adopting the scheme of this invention. In contrast, in the comparative experiment, the differences in charge levels among the battery clusters under the traditional method persisted, easily causing a single cluster to prematurely reach the full charge / discharge boundary, thus triggering a derating of the entire system. This invention achieves unbiased internal circulation balancing of the system through the zeroed-out power correction, thereby improving the usable capacity of the energy storage system.

Claims

1. A power allocation method for an energy storage system based on proportional-integral control, characterized in that, Includes the following steps: Collect the real-time state of charge of each battery cluster in the energy storage system and set the target state of charge; Calculate the state deviation between the real-time state of charge of each battery cluster and the target state of charge, and calculate the zeroed power correction amount based on the state deviation; Receive the total power scheduling instruction for the entire cabin, calculate the reference power corresponding to the total power scheduling instruction for the entire cabin, add the reference power and the zeroed power correction amount to obtain the preliminary power instruction, and when it is determined that the preliminary power instruction has an over-limit situation, allocate the over-limit power to obtain the final power instruction; The final power command is sent to the energy storage converter corresponding to each battery cluster for execution.

2. The power allocation method for an energy storage system based on proportional-integral control according to claim 1, characterized in that, The step of setting the target state of charge includes: Determine whether the energy storage system is in charging mode or discharging mode; In the charging mode, the real-time state of charge of all received battery clusters is traversed, and the real-time state of charge with the largest value is selected and set as the target state of charge. In the discharge mode, the real-time state of charge of all received battery clusters is traversed, and the real-time state of charge with the smallest value is selected and set as the target state of charge.

3. The power allocation method for an energy storage system based on proportional-integral control according to claim 2, characterized in that, The step of calculating the state deviation between the real-time state of charge of each battery cluster and the target state of charge includes: When it is determined that the energy storage system is in the charging mode, the target state of charge is subtracted from the real-time state of charge of each battery cluster to obtain the state deviation of the corresponding battery cluster. When the energy storage system is determined to be in the discharge mode, the target state of charge is subtracted from the real-time state of charge of each battery cluster to obtain the state deviation of the corresponding battery cluster.

4. The power allocation method for an energy storage system based on proportional-integral control according to claim 1, characterized in that, The step of calculating the zeroed power correction amount based on the state deviation includes: Using proportional-integral controllers corresponding to each battery cluster, the original power correction amount for the corresponding battery cluster is calculated based on the state deviation; Calculate the mean of the original power correction for all battery clusters; The zeroed power correction is obtained by subtracting the mean value from the original power correction.

5. The power allocation method for an energy storage system based on proportional-integral control according to claim 4, characterized in that, The step of using a proportional-integral controller corresponding to each battery cluster to calculate the original power correction amount for the corresponding battery cluster based on the state deviation includes: The state deviations corresponding to each battery cluster are respectively input into the corresponding proportional-integral controllers for mathematical calculations. The state deviation is received as an input quantity. The state deviation is multiplied by a proportional parameter using a proportional element to generate a power adjustment component. The state deviation is accumulated over time using an integral element and multiplied by an integral parameter to generate a power adjustment component. The power adjustment component output by the proportional element and the power adjustment component output by the integral element are added together to obtain the original power correction amount corresponding to each battery cluster.

6. The power allocation method for an energy storage system based on proportional-integral control according to claim 5, characterized in that, The step of calculating the reference power corresponding to the total cabin power scheduling command includes: Extract the absolute value of the total power scheduling command for the entire cabin as the initial demand; Collect the maximum power of each battery cluster; The maximum power of all battery clusters is summed to obtain the current total available maximum power of the system. Compare the initial demand with the current total available maximum power of the system. If the initial demand is greater than the current total available maximum power of the system, set the current total available maximum power of the system to the absolute value of the corresponding total power scheduling command for the entire cabin. When the initial demand is less than or equal to the current total available maximum power of the system, the initial demand is set to the absolute value of the total power scheduling command for the entire cabin. The absolute value of the total power scheduling command for the entire cabin is divided by the number of battery clusters to obtain the reference power allocated to an individual battery cluster.

7. The power allocation method for an energy storage system based on proportional-integral control according to claim 6, characterized in that, The step of allocating the excess power to obtain the final power command when it is determined that the initial power command exceeds the limit includes: Detect whether the preliminary power command is greater than the maximum power or less than 0; When the initial power command is greater than the maximum power, it is determined that the corresponding battery cluster has exceeded the upper limit. The initial power command is subtracted from the maximum power to obtain the over-limit power of the corresponding battery cluster, and the maximum power is set as the temporary command value after the corresponding battery cluster is truncated. When the initial power command is less than 0, it is determined that the corresponding battery cluster has exceeded the lower boundary limit. The initial power command is subtracted from 0 to obtain the over-limit power of the corresponding battery cluster, and 0 is set as the temporary command value after the corresponding battery cluster is truncated. When the initial power command is less than or equal to the maximum power and greater than or equal to 0, it is determined that the corresponding battery cluster is within the safe operating range and is a battery cluster that has not exceeded the limit. The over-limit power of the corresponding battery cluster is set to 0, and the initial power command is directly set to the temporary command value after the corresponding battery cluster is truncated.

8. The power allocation method for an energy storage system based on proportional-integral control according to claim 7, characterized in that, After obtaining the temporary instruction value, the following steps are also included: Extract the over-limit power corresponding to all battery clusters that have experienced the upper boundary over-limit situation and the lower boundary over-limit situation, and sum all the over-limit power to obtain the total over-limit power; Filter out battery clusters that have not exceeded the limit, extract the temporary instruction value corresponding to the battery clusters that have not exceeded the limit, and sum the temporary instruction values ​​corresponding to all battery clusters that have not exceeded the limit to obtain the total number of non-limit-exceeding instructions; Check if the sum of the non-limit-crossing instructions is equal to 0; If the sum of the non-limit-crossing instructions is equal to 0, further detect whether the number of non-limit-crossing battery clusters is greater than 0; If the number of battery clusters that have not exceeded the limit is greater than 0, the additional redistribution power is obtained by dividing the total over-limit power by the number of battery clusters that have not exceeded the limit. If the number of battery clusters that have not exceeded the limit is equal to 0, then the additional redistribution power is set to 0; If the sum of the non-over-limit instructions is greater than 0, calculate the allocation ratio corresponding to each non-over-limit battery cluster. The allocation ratio is equal to the value of the temporary instruction value of the corresponding non-over-limit battery cluster divided by the sum of the non-over-limit instructions. Multiply the total over-limit power by the allocation ratio to obtain the additional redistribution power. The temporary command value and the additional redistribution power are added together to obtain a new round of power command; The new round of power commands is used as the initial power command input for the limit detection step to perform loop boundary judgment until all commands of all battery clusters are between the maximum power and 0. When the loop stops, the power commands of all battery clusters that are within the safe operating range at the time of stopping the loop are uniformly set as the final power command.

9. The power allocation method for an energy storage system based on proportional-integral control according to claim 8, characterized in that, After the step of issuing the final power command to the energy storage converter corresponding to each battery cluster, the method further includes: The final power command is compared with the current maximum power of the corresponding battery cluster and 0 to determine their numerical equality. When the final power command is found to be equal to the maximum power or equal to 0, a saturation intervention signal is sent to the proportional-integral controller of the corresponding battery cluster. Upon receiving the saturation intervention signal, the state deviation of the corresponding battery cluster within the current sampling period is forcibly stopped from entering the integration stage for accumulation calculation, and the state deviation input to the proportional-integral controller is simultaneously forcibly set to zero.

10. A power distribution system for an energy storage system based on proportional-integral control, applied to the power distribution method for an energy storage system based on proportional-integral control as described in any one of claims 1-9, characterized in that, include: The status setting module is used to collect the real-time state of charge of each battery cluster in the energy storage system and set the target state of charge. The correction calculation module is used to calculate the state deviation between the real-time state of charge of each battery cluster and the target state of charge, and to calculate the zeroed power correction amount based on the state deviation. The over-limit allocation module is used to receive the total power scheduling instruction for the entire cabin, calculate the reference power corresponding to the total power scheduling instruction for the entire cabin, add the reference power and the zeroed power correction amount to obtain the preliminary power instruction, and when it is determined that the preliminary power instruction has an over-limit situation, allocate the over-limit power to obtain the final power instruction. The instruction execution module is used to send the final power instruction to the energy storage converter corresponding to each battery cluster for execution.

Images