Sodium-lithium hybrid energy storage system based on joint feasible power domain and control method
Patent Information
- Application Number
- CN202611548299.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-08
- Publication Date
- 2026-11-03
AI Technical Summary
Existing sodium-lithium hybrid energy storage systems suffer from cross-system state determination bias due to direct comparison of state of charge during state assessment, and lack a systematic feasible domain projection mechanism, leading to overcharge and over-discharge risks and power tracking deviations.
By constructing a control method based on a joint feasible power domain, the equivalent energy states of sodium and lithium sides are obtained, a joint feasible power domain is constructed, and the total power scheduling command is projected into the domain through a feasible domain projection module. The power control module performs power allocation, and truncation and deficit transfer are performed when the limit is exceeded. The return-to-center compensation module is used to keep the total grid connection point power unchanged.
It enables unified assessment of state of charge, avoids the risks of overcharging and over-discharging, reduces power point tracking deviation, and improves the safety and stability of sodium-lithium hybrid energy storage systems.
Smart Images

Figure CN123034484A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of hybrid energy storage control, and more particularly to a sodium-lithium hybrid energy storage system and control method based on a joint feasible power domain. BACKGROUND
[0002] With the complementary advantages of sodium ion batteries and lithium ion batteries gradually emerging in electrochemical characteristics, sodium-lithium hybrid energy storage collaborative control technology for large-scale energy storage power stations is particularly important. The sodium-lithium hybrid energy storage system connects the lithium branch and the sodium branch to the AC bus through independent bidirectional power converters, realizes the decoupled operation and unified scheduling of energy storage units of different voltage platforms, and is used to improve the power response capability and energy utilization efficiency under complex conditions.
[0003] Currently, the sodium-lithium hybrid energy storage system collects the temperature, state of charge and power margin of the sodium branch and the lithium branch, constructs a hybrid energy storage state mapping set, differentially allocates scheduling instructions to each branch, and uses independent thermal management linkage control to ensure operation safety, so as to effectively solve the voltage platform mismatch problem caused by direct mixing of sodium and lithium, and meet the basic hybrid energy storage power distribution demand.
[0004] However, in actual use, there are still some disadvantages, such as the existing scheme still bases the state evaluation of the sodium-lithium branch on the direct comparison of the respective state of charge, and the same state of charge value does not correspond to the same releasable energy, which leads to cross-system state determination deviation and easily causes overcharging and overdischarging or scheduling power cannot be fully executed.
[0005] At the same time, when the scheduling instruction exceeds the real-time feasible power interval of the branch, there is a lack of systematic feasible domain projection mechanism, the total power balance constraint and the real-time capability constraint on both sides are not coupled into a joint feasible domain and processed by truncation, shortage transfer and derating, which leads to the output instruction possibly exceeding the execution boundary of the converter or the battery management system, and the power tracking deviation significantly increases. SUMMARY
[0006] In order to overcome the above-mentioned defects of the prior art, the present application provides a sodium-lithium hybrid energy storage system and control method based on a joint feasible power domain, which solves the problems raised in the background art through the following scheme.
[0007] To achieve the above-mentioned purpose, the present application provides the following technical scheme:
[0008] The sodium-lithium hybrid energy storage system based on a joint feasible power domain comprises a sodium ion battery energy storage subsystem, a lithium ion battery energy storage subsystem, and bidirectional power converters connected to the sodium ion battery energy storage subsystem and the lithium ion battery energy storage subsystem respectively, the bidirectional power converters are connected to an AC bus, and the system further comprises a control device, the control device comprises:
[0009] State evaluation module: used for obtaining equivalent energy states of the sodium side and the lithium side, the equivalent energy states being determined according to corresponding side rated energy, health state and state of charge safety window;
[0010] Feasible region projection module: used for obtaining total power scheduling instructions, the total power scheduling instructions including a basic power component and a transient power component; constructing a joint feasible power domain according to real-time feasible power intervals of the sodium side and the lithium side and total power balance constraints; projecting initial power instructions corresponding to the total power scheduling instructions to the joint feasible power domain to obtain actual execution power instructions;
[0011] Power control module: used for controlling bidirectional power converters of the sodium side and the lithium side to perform power distribution according to the actual execution power instructions;
[0012] Return compensation module: used for calculating a return power according to deviations of the equivalent energy states of the sodium side and the lithium side from corresponding preset target energy states after the transient power component is executed; and controlling one side of the sodium side and the lithium side to increase or decrease discharge or increase or decrease charge by the return power and the other side to decrease or increase discharge or increase or decrease charge by an equal amount under the condition that total grid-connected point power is maintained unchanged.
[0013] To achieve the above object, the present application provides the following technical scheme: a sodium-lithium hybrid energy storage control method based on a joint feasible power domain, and a sodium-lithium hybrid energy storage system based on the joint feasible power domain, comprising:
[0014] S1: obtaining equivalent energy states of the sodium side and the lithium side, the equivalent energy states being determined according to corresponding side rated energy, health state and state of charge safety window, and used for representing releasable energy and / or chargeable energy of the corresponding side;
[0015] S2: obtaining total power scheduling instructions, the total power scheduling instructions including a basic power component and a transient power component;
[0016] S3: constructing a joint feasible power domain according to real-time feasible power intervals of the sodium side and the lithium side and total power balance constraints; projecting initial power instructions corresponding to the total power scheduling instructions to the joint feasible power domain to obtain actual execution power instructions; wherein, when the initial power instructions exceed the joint feasible power domain, performing truncation on the exceeding part, and transferring a deficiency generated by the truncation to the other side; if the other side still cannot meet the deficiency, performing at least one of derating, load shedding or alarming;
[0017] S4: controlling bidirectional power converters of the sodium side and the lithium side to perform power distribution according to the actual execution power instructions;
[0018] S5: After the transient power component is executed, calculate the return-to-center power based on the deviation between the equivalent energy state of the sodium side and the lithium side and the corresponding preset target energy state; while maintaining the total grid connection point power unchanged, control one side of the sodium side and the lithium side to increase or decrease the charge according to the return-to-center power, and control the other side to decrease or increase the charge by the same amount.
[0019] Preferably, in step S1, the equivalent energy state includes releasable energy in the discharge direction and rechargeable energy in the charging direction:
[0020] Energy can be released on the i-th side. ;
[0021] The i-th side can be charged with energy. ;
[0022] Where i∈{Na,Li}, Na represents the sodium side and Li represents the lithium side. Represented as the rated energy on the i-th side, This represents the health status of the i-th side. This represents the current charge state on the i-th side. This represents the lower limit of the safety window for the charged state on the i-th side. This represents the upper limit of the safety window for the charged state.
[0023] Preferably, the lower and upper limits of the state-of-charge safety window are determined based on at least one of the following: discharge cut-off voltage, charging cut-off voltage, rate capability, temperature boundary, and cycle life reduction factor of the corresponding cell.
[0024] Preferably, in step S2, the fundamental power component and the transient power component are obtained through low-pass filtering decomposition:
[0025]
[0026]
[0027] in, Represented as a low-pass filter operator, Represented as the fundamental power component, This is represented as a total power dispatch command. It is represented as a transient power component.
[0028] Preferably, step S3, determining the real-time feasible power range for the sodium side and the lithium side, specifically includes:
[0029] No. Upper limit of discharge in the real-time feasible power range: ;
[0030] Lower limit of charging direction: ;
[0031] Where i∈{Na,Li} , These represent the upper limits of discharge capacity and charging capacity of the i-th cell under the current SOC, temperature, and voltage conditions, respectively. , These are the discharge current limit value and the charging current limit value of the BMS on the i-th side, respectively; , These are the rated discharge capacity and rated charging capacity of the i-th bidirectional power converter, respectively.
[0032] Preferably, the joint feasible power domain Defined as:
[0033] ;
[0034] The projection is a boundary truncation mapping that maps the initial power command point located outside the joint feasible power domain to the point on the boundary of the joint feasible power domain that is closest to that point in Euclidean distance.
[0035] Preferably, the specific rules for S3, truncation and deficit transfer, are as follows:
[0036] When the initial power command on one side exceeds the real-time feasible power range on that side, the initial power value on that side is truncated to the corresponding boundary value, and the power deficit caused by the truncation is preferentially transferred to the other side.
[0037] If the remaining feasible power margin on the other side can cover the deficit, then the entire amount is transferred;
[0038] If only partial coverage is possible, the coverage will be moved to the boundary value on the other side, and the remaining uncovered amount will be subject to throttling, load shedding, or alarm processing.
[0039] Preferably, in step S3, the triggering conditions and execution order for derating, load shedding, or alarms are as follows:
[0040] When the deficit offset does not exceed the preset ratio of the total scheduling command, the derating is executed, and the basic power component and transient power component are reduced proportionally according to the ratio of the remaining feasible power margin on both sides.
[0041] If the minimum acceptable total power output cannot be met even after derating, load shedding will be performed.
[0042] At the same time, alarm information is reported to the energy management system. The alarm information includes the amount of deficit, the restricted side identifier, the trigger time, and the current real-time feasible power range on both sides.
[0043] The technical effects and advantages of this invention are as follows:
[0044] 1. This invention converts the state of charge (SOC) of the sodium and lithium sides, along with the rated energy, health status, and SOC safety window, into an equivalent energy state. It directly uses energy values to characterize the releaseable energy and / or chargeable energy of each side under the current conditions, thus unifying the SOC, which was originally not directly comparable, into an energy scale for evaluation. This avoids cross-system state determination bias caused by different releaseable energies corresponding to the same SOC value, reduces the risk of overcharging and over-discharging, and facilitates the full execution of scheduled power.
[0045] 2. The equivalent energy state is divided into releasable energy in the discharge direction and rechargeable energy in the charging direction. The releasable energy is obtained by multiplying the rated energy after health state deduction by the effective margin ratio of the current state of charge relative to the lower limit of the safety window. The rechargeable energy is obtained by multiplying the rated energy after health state deduction by the effective margin ratio of the upper limit of the safety window relative to the current state of charge. By incorporating both health state deduction and safety window margin into the quantification, the equivalent energy state can directly reflect the actual releasable and rechargeable capabilities of each side, thereby avoiding state judgment bias.
[0046] 3. This invention associates the safety window boundary with the actual operating boundary of the cell. Even if the sodium side and the lithium side are in the same state of charge, as long as the safety window range or health status of the two sides are different, the obtained equivalent energy state can reflect the difference in the actual safe and usable margin, thus avoiding overcharging and over-discharging caused by the state judgment not fully reflecting the cell safety boundary.
[0047] 4. This invention uses at least one of the ampere-hour integration method, internal resistance detection method, or capacity calibration method to estimate the health status, and triggers an update when the full charge completion event, full discharge completion event, or internal resistance growth rate exceeds a preset threshold; when battery aging causes the actual capacity to decrease, the equivalent energy state decreases accordingly, thereby tracking the change of the actual releaseable and chargeable energy on each side with the aging state, and avoiding the deviation from the fixed state of charge comparison method from increasing with aging.
[0048] 5. This invention further introduces a weighted fusion update of terminal voltage, temperature and internal resistance corrections, incorporating the deviation between measured operating parameters and calibrated reference values into the equivalent energy state correction; the corrected equivalent energy state is closer to the actual operating conditions corresponding to the current terminal voltage, temperature and internal resistance of the battery, reducing the problem of insufficient reflection of real-time operating conditions when calculated only by rated energy, state of health and state of charge safety window.
[0049] 6. Through progressive processing of anomaly retention, alarm and conservative contraction, the status assessment maintains continuity under abnormal data conditions. At the same time, the risks are limited by alarm and power boundary contraction, avoiding the increase of cross-system status misjudgment and overcharge and over-discharge risks due to input anomalies.
[0050] 7. The basic power component and transient power component are obtained by low-pass filtering decomposition. The basic power component smoothly tracks the continuous output demand, while the transient power component carries high-frequency step components. This separates the steady-state demand from the transient impact demand on the time scale, avoids the mixing of the two types of components when the dispatch command is directly issued, reduces power tracking error, and helps to allocate the transient power component to the side with higher capacity score, thereby reducing its continuous output pressure.
[0051] 8. When the initial power command exceeds the limit, the present invention first truncates it to the corresponding boundary, and the resulting deficit is preferentially transferred to the other side; if the remaining margin on the other side is insufficient, the remaining deficit is processed by derating, load shedding or alarm; ensuring that the actual executed power command is always within the joint feasible power domain, avoiding the output command from exceeding the execution boundary of the converter or battery management system, and reducing power tracking deviation;
[0052] 9. Handling remaining deficits through tiered processing: When the deficit is small, derating is performed, reducing the base power component and transient power component proportionally to the remaining feasible power margin on both sides; if the minimum acceptable total power output still cannot be met after derating, load shedding is performed; at the same time, alarm information is reported to the energy management system, including the number of deficits, the restricted side identifier, the trigger time, and the current real-time feasible power range on both sides; this enables the system to drate in an orderly manner when its capacity is insufficient, and to shed loads and issue alarms when necessary, avoiding loss of control at the execution boundary;
[0053] 10. This invention uses the equivalent energy state deviation as the basis for centering and maintains the total grid connection point power unchanged by adjusting both sides in equal and opposite ways, preventing the accumulation of energy state deviation on both sides after transient power distribution, and improving the safety and stability of sodium-lithium synergistic operation. Attached Figure Description
[0054] Figure 1 This is a general structural diagram of a sodium-lithium hybrid energy storage system based on a joint feasible power domain, according to an embodiment of this application.
[0055] Figure 2 This is a module connection diagram of the control device in a sodium-lithium hybrid energy storage system based on a joint feasible power domain, according to an embodiment of this application.
[0056] Figure 3 This is a flowchart of a sodium-lithium hybrid energy storage control method based on a joint feasible power domain, according to an embodiment of this application.
[0057] Figure 4 This is a flowchart of the equivalent energy state acquisition process according to an embodiment of this application.
[0058] Figure 5 This is a flowchart illustrating the breakdown of total power scheduling instructions provided according to an embodiment of this application.
[0059] Figure 6 This is a flowchart illustrating the execution of power control instructions provided in an embodiment of this application. Detailed Implementation
[0060] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0061] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to and includes any or all possible combinations of one or more of the listed items, and in the description of the embodiments of this application, unless otherwise stated, “a plurality” means two or more.
[0062] As attached Figure 1 The sodium-lithium hybrid energy storage system shown is based on the joint feasible power domain and includes a sodium-ion battery energy storage subsystem, a lithium-ion battery energy storage subsystem, a bidirectional power converter connected to the sodium-ion battery energy storage subsystem and the lithium-ion battery energy storage subsystem respectively, an AC bus connected to the AC side of the bidirectional power converter, and a control device.
[0063] The control device includes a state assessment module, a feasible region projection module, a power control module, and a centering-back compensation module, as shown in the attached diagram. Figure 2 As shown, where:
[0064] The state assessment module obtains the equivalent energy state of the sodium and lithium sides based on the corresponding side's rated energy, health state, and state of charge safety window.
[0065] The feasible domain projection module constructs a joint feasible power domain based on the real-time feasible power ranges of the sodium and lithium sides and the total power balance constraints, and projects the initial power command corresponding to the total power scheduling command onto the joint feasible power domain to generate the actual execution power command.
[0066] The power control module controls the bidirectional power converters on the sodium and lithium sides to perform power distribution according to the actual executed power command;
[0067] After the transient power component is executed, the return-to-center compensation module calculates the return-to-center power based on the deviation between the equivalent energy state of the sodium side and the lithium side and the corresponding preset target energy state, and controls the increase or decrease of discharge on one side and the decrease or increase of discharge on the other side by the same amount while keeping the total grid-connected power unchanged.
[0068] In one specific embodiment, it includes:
[0069] The sodium-ion battery energy storage subsystem is composed of sodium-ion battery cells connected in series and parallel. The DC output terminal of the sodium-ion battery energy storage subsystem is connected to the DC side port of the sodium-side bidirectional power converter.
[0070] The lithium-ion battery energy storage subsystem is composed of lithium-ion battery cells connected in series and parallel. The DC output terminal of the lithium-ion battery energy storage subsystem is connected to the DC side port of the lithium-side bidirectional power converter.
[0071] In the bidirectional power converter, at least one bidirectional power converter is configured on the sodium side. In an optional embodiment, the sodium-side bidirectional power converter consists of four PCS modules with a rated power of 100kW connected in parallel to form an equivalent PCS, which is then aggregated by a parallel coordination controller and connected to the AC bus; at least one bidirectional power converter is configured on the lithium side. The DC side of each bidirectional power converter is connected to the DC bus of the corresponding battery energy storage subsystem, and the AC side is connected to the AC bus. The bidirectional power converter supports bidirectional flow of active power, i.e., charging mode and discharging mode.
[0072] The AC bus connects the AC outputs of the sodium-side bidirectional power converter and the lithium-side bidirectional power converter. The AC bus is then connected to the power grid, microgrid, or industrial park distribution network via a grid connection point. A power sampling device is installed at the grid connection point to collect the total grid-connected power in real time.
[0073] The control device establishes a control command transmission link with the sodium-side bidirectional power converter and the lithium-side bidirectional power converter through one or more combinations of CAN bus, Modbus RTU, Modbus TCP or Ethernet; it collects the DC bus current, DC bus voltage and battery module temperature of the sodium-ion battery energy storage subsystem and the lithium-ion battery energy storage subsystem through current sensor sampling channel, voltage sensor sampling channel and temperature sensor sampling channel respectively; it receives the estimated state of charge value, estimated state of health value and real-time current limiting value from BMS through the communication interface, and receives the total power scheduling command.
[0074] The hardware carrier of the control device is at least one of DSP, FPGA, ARM processor, or industrial control computer. The output of the state evaluation module is connected to the input of the feasible region projection module and the input of the back-center compensation module, respectively; the output of the feasible region projection module is connected to the input of the power control module; the output of the power control module is connected to the control interface of the sodium-side bidirectional power converter and the lithium-side bidirectional power converter; the output of the back-center compensation module is connected to the power distribution adjustment interface of the power control module, or directly sends equal and opposite power adjustment commands to the sodium-side and lithium-side.
[0075] As attached Figure 3 The sodium-lithium hybrid energy storage control method based on the joint feasible power domain, as shown, is applied to the aforementioned sodium-lithium hybrid energy storage system based on the joint feasible power domain. The specific steps are as follows:
[0076] S1: Obtain the equivalent energy state of the sodium side and the lithium side. The equivalent energy state is determined based on the rated energy, health state and state of charge safety window of the corresponding side, and is used to characterize the energy that can be released and / or the energy that can be charged on the corresponding side.
[0077] S2: Obtain the total power scheduling instruction, which includes a basic power component and a transient power component;
[0078] S3: Based on the real-time feasible power ranges of the sodium and lithium sides and the total power balance constraints, a joint feasible power domain is constructed; the initial power command corresponding to the total power scheduling command is projected onto the joint feasible power domain to obtain the actual execution power command; wherein, when the initial power command exceeds the joint feasible power domain, the out-of-bounds portion is truncated, and the resulting deficit is transferred to the other side; if the other side still cannot meet the deficit, at least one of derating, load shedding, or alarm is executed;
[0079] S4: Control the bidirectional power converters on the sodium and lithium sides to perform power distribution according to the actual executed power command;
[0080] S5: After the transient power component is executed, calculate the return-to-center power based on the deviation between the equivalent energy state of the sodium side and the lithium side and the corresponding preset target energy state; while maintaining the total grid connection point power unchanged, control one side of the sodium side and the lithium side to increase or decrease the charge according to the return-to-center power, and control the other side to decrease or increase the charge by the same amount.
[0081] As attached Figure 4 As shown, S1 is executed by the state assessment module.
[0082] The input data required to obtain the equivalent energy state includes: the rated energy on the i-th side. Health status of the i-th side Current state of charge on the i-th side Lower limit of the safety window for the i-th side's charge state and the upper limit of the safety window for the state of charge Where i∈{Na,Li}, Na represents the sodium side and Li represents the lithium side; it also includes the energy that can be released on the i-th side. Energy can be charged into the i-th side. The sampling period for the above data is consistent with the communication period of the battery management system, with an example value of 100ms to 1s.
[0083] The total amount of energy that can actually be released or charged to the outside under the current health state and current state of charge can be obtained by multiplying the rated energy after health state deduction by the effective available margin ratio within the state of charge safety window: ; ;in, This represents the releasable state margin of the current state of charge on the i-th side relative to the lower limit of the safety window on that side; This represents the charge state margin of the current charge state on the i-th side relative to the upper limit of the safety window on that side.
[0084] The health status It is estimated by the corresponding battery management system using at least one of the following methods: ampere-hour integration method, internal resistance detection method, or capacity calibration method.
[0085] An update is triggered when the battery management system detects a full charge completion event or a full discharge completion event, or when the internal resistance growth rate exceeds a preset threshold. When the battery management system's health status estimate is unavailable, the previous valid health status value is retained for calculation. The preset threshold for the internal resistance growth rate can be calibrated by the battery management system manufacturer or determined from historical internal resistance change statistics.
[0086] The weighted fusion update incorporates corrections for terminal voltage, temperature, and internal resistance. The update formula is as follows: ,in, , , For the weighting coefficients, satisfying ; This is the measured value of the current open-circuit voltage on the i-th side; This is the reference value of the open-circuit voltage corresponding to the current state of charge of the i-th side under calibration conditions; The current temperature on the i-th side; For calibration temperature; This is the measured value of the current DC internal resistance on the i-th side; This is the reference value for DC internal resistance under calibration conditions. Corrections for chargeable energy can be made in the same manner. Replace with implement.
[0087] Alternatively, a Kalman filter fusion update method can be adopted, using the state of charge estimated by the battery management system as the state variable and the measured terminal voltage as the observation. The estimated state of charge is iteratively updated by extended Kalman filter, and then substituted into the equivalent energy state calculation formula to obtain the corrected equivalent energy state.
[0088] Both of the above-mentioned fusion update methods are preferred implementation methods for obtaining equivalent energy states.
[0089] When any input data required for the equivalent energy state calculation is missing or abnormal, the state assessment module retains the equivalent energy state calculation value from the previous valid cycle and issues an alarm signal; if the abnormality persists for more than a preset time, the corresponding real-time feasible power range is narrowed to a conservative preset value. The preset time is 5 seconds as an example.
[0090] As attached Figure 5 As shown, S2 is executed by the feasible region projection module.
[0091] Total power dispatch command The source of the total power dispatch command is the power grid dispatch center's AGC system, EMS energy management system, or microgrid central controller, and the command format is a periodically updated power command value. The update cycle of the total power dispatch command is 1s to 5s (AGC frequency regulation scenario) or 5s to 60s (peak shaving and valley filling scenario). The feasible region projection module receives the total power dispatch command through the communication interface of the control device.
[0092] Fundamental power component The component in the total power dispatch command that carries the steady-state and continuous power demand of the system is the transient power component. In this embodiment, the time scale is a continuous output range of more than 1 second to several minutes. The component in the total power dispatch command that carries the system's transient and impulsive power demand is the time scale in this embodiment, which is the transient output range from sub-second to several seconds.
[0093] The acquisition methods for the basic power component and transient power component include: the EMS or AGC system directly issuing the basic power component and transient power component separately; when the total power scheduling command uses a single total power value... When the data is distributed, the decomposition operation is performed by the feasible domain projection module or a specially designed decomposition stage.
[0094] In this embodiment, when the total power scheduling command uses a single total power value When the data is distributed, the feasible domain projection module or a specially designed decomposition process performs a decomposition operation to obtain the basic power components and transient power components for description.
[0095] In an optional embodiment, the basic decomposition method employs a low-pass filter: , , This is a low-pass filter operator with a cutoff frequency of . It is determined based on the spectral characteristics of the power signal.
[0096] The cutoff frequency is used to ensure that the basic power component smoothly tracks the continuous output demand in the total power dispatch command, and to separate the high-frequency step components into transient power components.
[0097] In digital control devices, the low-pass filter can be a first-order low-pass filter transfer function. Discretization implementation, The corresponding discrete difference equation is expressed as: ,in, To control the cycle, This is the sampling sequence number.
[0098] In a preferred embodiment, the cutoff frequency of the low-pass filter is dynamically adjusted based on the real-time capability scores of the sodium-side and lithium-side. (Corresponding capability scores) The calculation is expressed as: , Score the ability of the i-th side; This represents the current available power on this side; This is the rated power on that side; This represents the current actual power ramp-up rate on this side; This is the reference power ramp rate on this side; This is the current per-unit value of the DC internal resistance on this side; This represents the current charge / discharge efficiency on this side. This is a dimensionless aging cost score, ranging from 0 to 1, which can be calculated from the number of battery cycles and the initial capacity decay rate on that side. For the weighting coefficients, satisfying .
[0099] All of the above weight coefficients are non-negative, and the corresponding terms have been normalized.
[0100] Transient power components are preferentially assigned to the side with higher capability scores, and the decomposition cutoff frequency is dynamically adjusted according to the capability score of the side responsible for the transient: the higher the capability score, the higher the cutoff frequency, allowing the transient components to contain higher frequency components.
[0101] In an alternative embodiment, the cutoff frequency can be determined using a linear mapping relationship: ,in, and These are the lower and upper limits of the preset cutoff frequency, respectively, with example values of 0.05Hz and 0.2Hz; This is a capability score for handling transient power components. The above mapping relationship means that the higher the capability score, the wider the allowed transient bandwidth.
[0102] In one specific embodiment, when the total power scheduling command exist At 100kW, The clock speed jumps to 200kW, remains there for 10 seconds, and then drops back to 100kW. A first-order low-pass filter is used, with a cutoff frequency of... Corresponding time constant ≈1.59s. Therefore, after the step jump, the increment of the fundamental power component is: ;
[0103] exist At that time, the increment of the fundamental power component is The basic power component value is 146.6kW; the transient power component value is 200-146.6=53.4kW.
[0104] exist At that time, the increment of the fundamental power component is The basic power component value is 195.7kW; the transient power component value is 4.3kW.
[0105] exist At that time, the increment of the basic power component reached 99.8kW, the basic power component value was 199.8kW, and the transient power component value was approximately 0.2kW. Low-pass filtering decomposition enabled the basic power component to smoothly track the continuous power output demand, and the transient power component only had a significant amplitude in the first few seconds, achieving effective separation of high-frequency step components.
[0106] If an adaptive cutoff frequency is used, and a sodium-side capability score is set... Lithium-side capability rating The cutoff frequency is then determined based on the sodium-side ability score. Corresponding time constant .exist At that time, the increment of the fundamental power component is The base power component is 171.6kW, and the transient power component is 28.4kW. After the adaptive cutoff frequency is increased, the transient power component decays faster, and the base power component can meet the continuous demand more quickly. This reduces the continuous output pressure on the transient side and helps protect transient response resources.
[0107] S3 is executed by the feasible region projection module.
[0108] The feasible domain projection module constructs a joint feasible power domain based on the real-time feasible power ranges of the sodium and lithium sides and the total power balance constraints. It projects the initial power command to the joint feasible power domain, generates the actual power command to be executed, and performs truncation, shortfall transfer, derating, load shedding or alarm processing for out-of-bounds situations.
[0109] Determine the real-time feasible power range for both the sodium and lithium sides.
[0110] The upper limit of discharge and the lower limit of charging direction for the real-time feasible power range on the i-th side are respectively: ; , where i∈{Na,Li}; , These represent the upper limits of discharge capacity and charging capacity of the i-th cell under the current SOC, temperature, and voltage conditions, respectively. , These are the discharge current limit value and the charging current limit value of the BMS on the i-th side, respectively; , These are the rated discharge capacity and rated charging capacity of the i-th bidirectional power converter, respectively, both being positive values on the AC side.
[0111] If the cell capacity and BMS current limit values are given on the DC side, they should be converted to the AC side based on the bidirectional power converter efficiency before being compared. The real-time feasible power range is the executable boundary after the cell capacity is constrained by both the BMS protection threshold and the rated capacity of the bidirectional power converter.
[0112] The total power balance constraint, neglecting bidirectional power converter losses, is: When considering the losses of the bidirectional power converter, the correction is as follows:
[0113] Discharge conditions: ;
[0114] Charging conditions: ;
[0115] in, The power at the DC terminal of the i-th battery is... This represents the actual output power of the i-th AC side; , These figures represent the conversion efficiencies of the sodium-side and lithium-side bidirectional power converters under corresponding operating conditions. All power values are signed, with positive for discharging and negative for charging.
[0116] Furthermore, the joint feasible power domain AC power is defined as follows: .
[0117] When considering converter losses, the feasible range on the DC side can be first converted into a feasible range on the AC side, and then a joint feasible power domain can be established according to the above formula. The joint feasible power domain is a line segment, point, or empty set determined by the intersection of the real-time feasible power range on the sodium side, the real-time feasible power range on the lithium side, and the total power balance constraint in a two-dimensional plane.
[0118] The criterion for determining a point within a domain is: if a point... If all three constraints are met, the point is located within the joint feasible power domain; otherwise, the point is located outside the joint feasible power domain.
[0119] The initial power command is determined according to the preset allocation strategy.
[0120] Let i∈{Na,Li} denote the i-th side, and assume... For the side that bears the transient power component, then: Among them, when , ;when hour, ; The side with the higher capability score in S2 is selected; when the capability scores of both sides are equal, the sodium side is assumed to bear the transient power component by default. The distribution coefficients of the basic power components satisfy... The value is determined based on the rated energy ratio of the sodium side to the lithium side, energy cost, or operating strategy.
[0121] The specific rules for boundary violation detection and truncation are as follows: Check the initial power command point. Is it located in the joint feasible power domain? Inside.
[0122] If a point satisfies all three constraints, it is determined to be an in-domain point and does not require correction; it is directly used as the actual power instruction to be executed. If a point does not satisfy any of the constraints, it is determined to be out of bounds.
[0123] When the i-th side exceeds the limit, the initial power value of the i-th side is truncated: if Then it is truncated to ;like Then it is truncated to The power deficit resulting from the cutoff is: The cutoff rules in the charging direction are symmetrical, and the constraint relationship is simply replaced with the corresponding interval in the charging direction.
[0124] The specific rules for deficit transfer are as follows: when a power deficit is generated on one side due to insufficient capacity, the deficit is preferentially transferred to the other side, provided that the current remaining feasible power margin on the other side can cover the deficit.
[0125] Taking the discharge direction as an example, when the sodium side is cut off and a deficit is generated... At that time, if the residual margin in the discharge direction of the lithium side satisfies The shortfall will then be fully transferred to the lithium side, i.e. ;like The shortfall will then be transferred, i.e. The remaining uncovered shortfall The system will initiate derating, load shedding, or alarm processing. The rules for transferring insufficient charge in the charging direction are symmetrical.
[0126] The triggering conditions and execution order for derating, load shedding, or alarms are as follows: if the remaining deficit cannot be met on the other side after the deficit is transferred, at least one of derating, load shedding, or alarms will be executed.
[0127] The derating is triggered when the deficit offset does not exceed a preset percentage of the total dispatch command. The preset percentage is typically 10% to 20% of the total dispatch command amplitude, and the specific value is set by the EMS or AGC system based on the acceptable power deviation range. The derating is executed as follows: the maximum allowable total AC power on both sides is set to... ;like Then the total grid connection point power command will be reduced to , and according to The basic power components are redistributed, and the transient power components are still handled by the transient side before the feasible domain verification is performed. If the limit is still exceeded, the upper limit of the real-time feasible power range on both sides (or the lower limit of the charging direction) is directly taken as the actual power command to be executed, and the remaining unmet part enters the load shedding or alarm.
[0128] If the minimum acceptable total power output cannot be met after derating, load shedding will be performed, cutting off the lowest priority load branch among the interruptible loads; while performing derating or load shedding, alarm information will be reported to the EMS, AGC or microgrid central controller. The alarm information includes the amount of deficit, the restricted side identifier, the trigger time, and the current real-time feasible power range on both sides.
[0129] The minimum acceptable total power output can be set by the upper-level EMS or AGC system based on system frequency, voltage safety requirements, or interruptible load constraints.
[0130] In a preferred embodiment, the projection uses the least squares projection criterion instead of the boundary truncation mapping, mapping the initial power command point to the point within the joint feasible power domain that minimizes the sum of the squares of the two-sided power adjustments: .
[0131] In another preferred embodiment, the projection adopts a capacity margin weighted projection, which determines the allocation weight of the out-of-bounds correction amount according to the ratio of the current remaining feasible power margin on the sodium side and the lithium side, so that the side with the larger margin can bear more of the deficit transfer.
[0132] As attached Figure 6 As shown, S4 is executed by the power control module.
[0133] The power control module receives the actual execution power command output by the feasible domain projection module, converts it into the active power setpoint of the sodium-side bidirectional power converter and the lithium-side bidirectional power converter, and sends it to the power loop controller of the corresponding bidirectional power converter. The local controller of the bidirectional power converter completes the underlying power tracking closed loop to realize the actual power allocation between the sodium side and the lithium side.
[0134] Receive the actual execution power command output by the feasible region projection module The active power setpoint of the sodium-side bidirectional power converter is set to... The active power setpoint of the lithium-side bidirectional power converter is set to .
[0135] Control commands are issued to the bidirectional power converter in two ways: either the power loop setpoint method, which sends the target active power value to the power loop controller of the bidirectional power converter; or the current loop setpoint method, which sends the target current value to the inner current loop of the bidirectional power converter.
[0136] This embodiment uses the power loop provisioning method as an example for illustration.
[0137] The local controller of the bidirectional power converter adopts a cascaded control structure of voltage outer loop and current inner loop. In power loop setpoint mode, the target active power value output by the power control module is first converted into the d-axis current setpoint value of the current inner loop via the power outer loop or directly.
[0138] Target active power in a synchronously rotating coordinate system With d-axis current setpoint The relationship is: ,in, Let be the d-axis component of the grid connection point voltage in a synchronous rotating coordinate system. When When W is the unit, the above formula holds directly; when When the unit is kW, it should be written as In this embodiment, the power setpoint is uniformly expressed in kW, and the conversion relationship is completed by the control device when it is sent to the inner current loop.
[0139] when When negative, it indicates charging condition; the bidirectional power converter operates in charging mode, and the d-axis current setpoint of the inner current loop is correspondingly reversed. When the voltage is close to zero, the current can be switched to be output from the outer voltage loop to avoid division by zero.
[0140] The inner current loop performs closed-loop regulation on the d-axis current and q-axis current respectively. Its output is transformed into a modulation wave and then used to generate a switching drive signal through space vector pulse width modulation (SVPWM) or sinusoidal pulse width modulation (SPWM) to control the phase and amplitude of the output current of the bidirectional power converter, so that the actual output active power tracks the target active power value.
[0141] The inner current loop uses a proportional-integral regulator or a proportional-resonant regulator, and its control parameters are determined based on the filter inductance, switching frequency, and desired bandwidth of the bidirectional power converter.
[0142] In one specific embodiment, the inner current loop bandwidth is set to 1 / 10 to 1 / 20 of the switching frequency. When the switching frequency is 5kHz, the exemplary value of the inner current loop bandwidth is 250 to 500Hz. The outer voltage loop bandwidth is set to 1 / 5 to 1 / 10 of the inner current loop bandwidth to achieve stability of the cascaded control.
[0143] In an implementation where four PCS modules with a rated power of 100kW are connected in parallel on the sodium side to form an equivalent PCS, the power control module sends the target active power value on the sodium side to the parallel coordination controller. The parallel coordination controller decomposes the target active power value into the power setpoint values of each PCS module according to the capacity ratio or equal distribution method, and ensures that the AC side outputs of each module are synchronized in parallel. The local controller of each PCS module executes the above-mentioned cascaded control and pulse width modulation to achieve parallel current sharing and total power point tracking.
[0144] When the power control module performs power allocation, it causes a change in the equivalent energy state of the sodium and lithium sides. The changed equivalent energy state is sampled and updated by the state evaluation module and then enters the return-to-center compensation module to provide input for S5.
[0145] S5 is executed by the return-to-center compensation module.
[0146] After the transient power component is executed, the return-to-center compensation module reads the updated equivalent energy states of the sodium and lithium sides provided by the state assessment module, compares them with the preset target energy state to calculate the deviation, generates the return-to-center power, and performs complementary return-to-center action while maintaining the total grid-connected power unchanged, so that the side that deviates from the target energy state is restored to the target value.
[0147] The trigger condition for return-to-center compensation is the completion of transient power component execution.
[0148] The determination of the completion of the transient power component is made by at least one of the following methods: the time window of the transient power component ends, that is, the preset transient duration expires; the amplitude of the transient power component decays to below a preset threshold, for example, when the transient power amplitude is less than 5% of the basic power component, it is determined that the transient has ended; or the transient flag bit in the total power scheduling command changes from valid to invalid.
[0149] After triggering the return-to-center compensation, the return-to-center compensation module obtains the updated actual values of the equivalent energy state on the sodium and lithium sides from the state assessment module. and And read the preset target energy state from the storage unit. and .
[0150] The equivalent energy state deviations on the sodium and lithium sides are calculated as follows: Among them, the preset target energy state The setting is based on at least one of the following: the equivalent energy state value corresponding to the midpoint of the corresponding side's charge state safety window; the equivalent energy state value of the corresponding side at the system initialization time; and the target energy weight allocation value set according to the operation strategy.
[0151] Deviation value When the value is positive, it indicates that the actual equivalent energy state on that side is higher than the target value, and the energy on that side needs to be reduced; when the deviation value is negative, it indicates that the actual equivalent energy state on that side is lower than the target value, and the energy on that side needs to be increased.
[0152] Under discharge conditions, increasing discharge lowers the equivalent state of energy (ESE) on that side, while decreasing discharge raises it; under charging conditions, increasing charging raises the ESE on that side, while decreasing charging lowers it. At that time, discharge is increased during discharge conditions and charging is reduced during charging conditions; when During this time, discharge is reduced under discharge conditions and charging is increased under charging conditions.
[0153] The calculation of the return power is expressed as follows: , , This is the gain proportional to the centering ratio; This is the saturation limiting function, and its value is: when hour, ;when hour, ;when hour, .
[0154] in, The maximum amplitude limit of the return-to-center power is within a preset proportion of the rated capacity of the dual-sided bidirectional power converter, or a preset power change rate limit multiplied by the return-to-center execution time window. Return-to-center proportional gain. The value of is determined based on the reciprocal of the execution time window during the return period; in this embodiment, the value is . ,in This is the expected return-to-center completion time constant.
[0155] The centering compensation module calculates the result. Then, the feasible power range is verified in real time in S3: if If the sum of the current S4 actual executed power command exceeds the real-time feasible power range on the i-th side, then... The amplitude is limited, and the mismatch caused by the amplitude limit is transferred to the preset reduction adjustment or waits for the next control cycle to return to the center.
[0156] The complementary back-to-center execution mode under total power preservation is as follows: During the back-to-center compensation execution process, the total grid-connected power output by the system remains equal to the current scheduling requirement, i.e., it satisfies: , , These are the power adjustment amounts for the sodium side and the lithium side, respectively.
[0157] The specific execution correspondence in the complementary return is as follows: Under discharge conditions, when the equivalent energy state of the first side is higher than the target value ( > When ), add to the first side The second side is reduced by the same amount. When the equivalent energy state on the first side is lower than the target value ( When <0), the amount of material on the first side is reduced, and the amount of material on the second side is increased by the same amount.
[0158] During charging, when the equivalent energy state of the first side is higher than the target value, the first side is charged less (charging power is reduced), and the second side is charged more by the same amount (charging power is increased); when the equivalent energy state of the first side is lower than the target value, the first side is charged more, and the second side is charged less by the same amount.
[0159] The power adjustment amounts on both sides are equal in magnitude and opposite in direction, so that the total grid connection point power remains unchanged.
[0160] When the return-to-center compensation module executes through the power distribution adjustment interface of the power control module, Superimposed on the current sodium-side active power given, - This is superimposed on the current lithium-side active power setpoint to form new sodium-side and lithium-side power setpoint values; when directly transmitting, the power is transmitted to the sodium side. and issued to the lithium side The two sides still satisfy the condition after being superimposed. .
[0161] Central proportional gain Segmented settings can be adopted: when When the gain is large, a larger gain is used to accelerate the return to center; when When the gain is small but not yet zero, a smaller gain is used to reduce the impact of the return-to-neutral on the smoothness of the total power distribution.
[0162] The return-to-center power has an additional power change rate limit: , Determined based on the allowable power ramp-up rate of the bidirectional power converter.
[0163] Secondly: The accompanying drawings of the embodiments disclosed in this invention only involve the structures involved in the embodiments disclosed in this invention. Other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other.
[0164] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A sodium-lithium hybrid energy storage system based on a combined feasible power domain, comprising a sodium-ion battery energy storage subsystem, a lithium-ion battery energy storage subsystem, and a bidirectional power converter connected to the sodium-ion battery energy storage subsystem and the lithium-ion battery energy storage subsystem respectively, wherein the bidirectional power converter is connected to an AC bus, characterized in that, It also includes a control device, which comprises: State assessment module: used to obtain the equivalent energy state of the sodium side and lithium side, the equivalent energy state being determined based on the corresponding side's rated energy, health state, and state of charge safety window; Feasible domain projection module: used to obtain the total power scheduling instruction, which includes a basic power component and a transient power component; construct a joint feasible power domain based on the real-time feasible power ranges of the sodium side and the lithium side and the total power balance constraints; project the initial power instruction corresponding to the total power scheduling instruction onto the joint feasible power domain to obtain the actual executed power instruction; Power control module: used to control the bidirectional power converter on the sodium side and lithium side to perform power distribution according to the actual executed power command; The return-to-center compensation module is used to calculate the return-to-center power based on the deviation between the equivalent energy state of the sodium side and the lithium side and the corresponding preset target energy state after the transient power component is executed; and to control one side of the sodium side and the lithium side to increase or decrease the charge according to the return-to-center power while keeping the total grid-connected power constant, and to decrease or increase the charge on the other side by the same amount.
2. A sodium-lithium hybrid energy storage control method based on a joint feasible power domain, applied to the sodium-lithium hybrid energy storage system based on a joint feasible power domain as described in claim 1, characterized in that, include: S1: Obtain the equivalent energy state of the sodium side and the lithium side. The equivalent energy state is determined based on the rated energy, health state and state of charge safety window of the corresponding side, and is used to characterize the energy that can be released and / or the energy that can be charged on the corresponding side. S2: Obtain the total power scheduling instruction, which includes a basic power component and a transient power component; S3: Based on the real-time feasible power ranges of the sodium and lithium sides and the total power balance constraints, a joint feasible power domain is constructed; the initial power command corresponding to the total power scheduling command is projected onto the joint feasible power domain to obtain the actual execution power command; wherein, when the initial power command exceeds the joint feasible power domain, the out-of-bounds portion is truncated, and the resulting deficit is transferred to the other side; if the other side still cannot meet the deficit, at least one of derating, load shedding, or alarm is executed; S4: Control the bidirectional power converters on the sodium and lithium sides to perform power distribution according to the actual executed power command; S5: After the transient power component is executed, calculate the return-to-center power based on the deviation between the equivalent energy state of the sodium side and the lithium side and the corresponding preset target energy state; while maintaining the total grid connection point power unchanged, control one side of the sodium side and the lithium side to increase or decrease the charge according to the return-to-center power, and control the other side to decrease or increase the charge by the same amount.
3. The sodium-lithium hybrid energy storage control method based on the joint feasible power domain according to claim 2, characterized in that: In S1, the equivalent energy state includes releasable energy in the discharge direction and rechargeable energy in the charging direction: Energy can be released on the i-th side. ; The i-th side can be charged with energy. ; Where i∈{Na,Li}, Na represents the sodium side and Li represents the lithium side. Represented as the rated energy on the i-th side, This represents the health status of the i-th side. This represents the current charge state on the i-th side. This represents the lower limit of the safety window for the charged state on the i-th side. This represents the upper limit of the safety window for the charged state.
4. The sodium-lithium hybrid energy storage control method based on the joint feasible power domain according to claim 3, characterized in that: The lower and upper limits of the state-of-charge safety window are determined based on at least one of the following: discharge cut-off voltage, charging cut-off voltage, rate capability, temperature boundary, and cycle life reduction factor of the corresponding cell.
5. The sodium-lithium hybrid energy storage control method based on the joint feasible power domain according to claim 2, characterized in that: In step S2, the fundamental power component and the transient power component are obtained through low-pass filtering decomposition: in, Represented as a low-pass filter operator, Represented as the fundamental power component, This is represented as a total power dispatch command. It is represented as a transient power component.
6. The sodium-lithium hybrid energy storage control method based on the joint feasible power domain according to claim 2, characterized in that: S3, determining the real-time feasible power range for the sodium side and the lithium side, specifically includes: No. Upper limit of discharge in the real-time feasible power range: ; Lower limit of charging direction: ; Where i∈{Na,Li} , These represent the upper limits of discharge capacity and charging capacity of the i-th cell under the current SOC, temperature, and voltage conditions, respectively. , These are the discharge current limit value and the charging current limit value of the BMS on the i-th side, respectively; , These are the rated discharge capacity and rated charging capacity of the i-th bidirectional power converter, respectively.
7. The sodium-lithium hybrid energy storage control method based on the joint feasible power domain according to claim 2, characterized in that: The joint feasible power domain Defined as: ; The projection is a boundary truncation mapping that maps the initial power command point located outside the joint feasible power domain to the point on the boundary of the joint feasible power domain that is closest to that point in Euclidean distance.
8. The sodium-lithium hybrid energy storage control method based on the joint feasible power domain according to claim 2, characterized in that: The specific rules for S3, truncation, and deficit transfer are as follows: When the initial power command on one side exceeds the real-time feasible power range on that side, the initial power value on that side is truncated to the corresponding boundary value, and the power deficit caused by the truncation is preferentially transferred to the other side. If the remaining feasible power margin on the other side can cover the deficit, then the entire amount is transferred; If only partial coverage is possible, the coverage will be moved to the boundary value on the other side, and the remaining uncovered amount will be subject to throttling, load shedding, or alarm processing.
9. The sodium-lithium hybrid energy storage control method based on the joint feasible power domain according to claim 2, characterized in that: In S3, the triggering conditions and execution order for derating, load shedding, or alarms are as follows: When the deficit offset does not exceed the preset ratio of the total scheduling command, the derating is executed, and the basic power component and transient power component are reduced proportionally according to the ratio of the remaining feasible power margin on both sides. If the minimum acceptable total power output cannot be met even after derating, load shedding will be performed. At the same time, alarm information is reported to the energy management system. The alarm information includes the amount of deficit, the restricted side identifier, the trigger time, and the current real-time feasible power range on both sides.
