Energy storage pulse collaborative control method and system based on electrochemical recovery

Abstract

The present application relates to the technical field of electrochemical energy storage system control, and particularly relates to an energy storage pulse collaborative control method and system based on electrochemical recovery; the method comprises the following steps: obtaining pulse characteristic parameters of a load through a high-frequency data acquisition device; constructing an electrochemical relaxation time constant model; calculating an effective recovery window of a battery and generating a time-domain matching result; in response to the time-domain matching result representing that a natural downtime is less than the effective recovery window, dividing a plurality of virtual subsets in a logic layer; constructing a multi-objective collaborative scheduling model to obtain an optimal control sequence for completing energy storage pulse collaborative control. In this way, the technical problem of lack of control strategy adaptability of existing electrochemical energy storage life management technology under high-frequency pulse working conditions is solved, and the dynamic response adaptability, operation economy and battery capacity release efficiency of the energy storage system under pulse load scenarios are improved.

Description

Technical Field

[0001] This invention relates to the field of electrochemical energy storage system control technology, and in particular to a method and system for coordinated control of energy storage pulses based on electrochemical recovery. Background Technology

[0002] With the global energy transition, user-side electrochemical energy storage systems, as a key technology for improving grid flexibility and realizing the local consumption of distributed energy, are widely used in typical industrial and commercial scenarios such as port cranes, industrial spot welding, and precision machining. However, the loads in these scenarios often exhibit high-frequency, high-rate, and intermittent pulse impact characteristics, posing severe challenges to the instantaneous power response capability and long-term cycle life of energy storage systems. To cope with the power fluctuations caused by pulse loads, existing technologies mostly adopt solutions such as capacity over-provisioning or hybrid energy storage systems (HESS) that couple batteries and supercapacitors. The former leads to a significant increase in the total life cycle cost due to hardware redundancy, while the latter, although alleviating power pressure, introduces topological complexity and difficulties in energy management and maintenance.

[0003] Furthermore, in battery life management, existing technologies have developed various methods for quantifying battery life degradation. For example, prior art document 1 (application publication number CN119004866B) discloses an electrochemical energy storage life degradation quantification system. This system collects multi-dimensional data such as operating temperature and current density, and combines impedance monitoring with a capacity degradation prediction model to quantitatively assess and warn of battery life degradation trends. While this system can reflect the battery aging state to some extent, its model is mainly based on static or quasi-static operating conditions and fails to fully consider the nonlinear dynamic characteristics of the internal electrochemical processes of the battery under pulsed loads, especially the accumulation of polarization effects caused by high-current discharge and the recovery effect during the pause phase. Specifically, under continuous high-frequency pulse impacts, the battery terminal voltage drops rapidly to the cutoff threshold due to ion diffusion kinetics limitations, causing a "pseudo-empty battery" phenomenon. Existing life degradation quantification systems lack adaptability to such transient polarization conditions, and their prediction models still rely on the linear scheduling of macroscopic parameters, failing to accurately capture the impact of microscopic electrochemical relaxation mechanisms on life degradation. This makes it difficult for existing technologies to achieve synergistic optimization of battery effective capacity release rate and cycle life under pulsed load scenarios, often resulting in problems such as premature voltage cutoff, low capacity utilization, and exacerbation of irreversible side reactions. Therefore, existing electrochemical energy storage life management technologies suffer from a lack of adaptability in control strategies under high-frequency pulsed operating conditions. Summary of the Invention

[0004] To address the aforementioned shortcomings or deficiencies, this invention provides a method and system for coordinated control of energy storage pulses based on electrochemical recovery, which can solve the technical problem of insufficient adaptability of control strategies in existing electrochemical energy storage lifetime management technologies under high-frequency pulse conditions.

[0005] This invention provides a method for coordinated control of energy storage pulses based on electrochemical recovery, comprising:

[0006] The pulse characteristic parameters of the load are obtained through a high-frequency data acquisition device.

[0007] An electrochemical relaxation time constant model was constructed based on pulse characteristic parameters and the principles of electrochemical kinetics.

[0008] The battery operating parameters are input into the electrochemical relaxation time constant model. The effective recovery window of the battery is calculated through the electrochemical relaxation time constant model. The natural pause time in the pulse characteristic parameters is compared with the effective recovery window to generate time-domain matching results.

[0009] In response to the time-domain matching result indicating that the natural pause time is less than the effective recovery window, the energy storage battery cluster is divided into multiple virtual subsets in the logic control layer of the energy management system, and at least one virtual subset is maintained in a discharge state at any given time.

[0010] A multi-objective cooperative scheduling model is constructed with the objectives of minimizing the cumulative active voltage and maximizing the effective discharge capacity as objective functions, and the optimal control sequence of the multi-objective cooperative scheduling model is obtained by using an optimization algorithm.

[0011] Based on the optimal control sequence, the power allocation and state switching operations of each virtual subset are executed through the power converter to complete the energy storage pulse coordinated control.

[0012] According to a second aspect, this invention provides an energy storage pulse coordinated control system based on electrochemical recovery, comprising:

[0013] The pulse characteristic parameter acquisition module is used to acquire the pulse characteristic parameters of the load through a high-frequency data acquisition device.

[0014] The time constant model construction module is used to construct an electrochemical relaxation time constant model based on pulse characteristic parameters and the principles of electrochemical kinetics.

[0015] The time-domain matching result generation module is used to input battery operating parameters into the electrochemical relaxation time constant model, calculate the effective recovery window of the battery through the electrochemical relaxation time constant model, and compare the natural pause time in the pulse characteristic parameters with the effective recovery window to generate time-domain matching results.

[0016] The energy storage virtual subset establishment module is used to divide the energy storage battery cluster into multiple virtual subsets in the logic control layer of the energy management system when the natural pause time of the time domain matching result is less than the effective recovery window, and to maintain that at least one virtual subset is in a discharge state at any given time.

[0017] The optimal control sequence solving module is used to construct a multi-objective cooperative scheduling model with the objective functions of minimizing the cumulative active voltage and maximizing the effective discharge capacity, and to use an optimization algorithm to solve for the optimal control sequence of the multi-objective cooperative scheduling model.

[0018] The energy storage pulse coordinated control module is used to perform power allocation and state switching operations of each virtual subset through the power converter according to the optimal control sequence to complete the energy storage pulse coordinated control.

[0019] According to a third aspect, the present invention provides an electronic device comprising:

[0020] At least one processor; and a memory communicatively connected to the at least one processor;

[0021] The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform any of the electrochemical recovery-based energy storage pulse collaborative control methods in the embodiments of the present invention.

[0022] According to another aspect of the present invention, a non-transient computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause a computer to execute any of the electrochemical recovery-based energy storage pulse coordinated control methods in the embodiments of the present invention.

[0023] The present invention provides a method for coordinated control of energy storage pulses based on electrochemical recovery. This method is achieved through six core steps: obtaining pulse characteristic parameters, constructing an electrochemical relaxation time constant model, generating time-domain matching results, partitioning logical virtual subsets, solving a multi-objective coordinated scheduling model, and executing power allocation. Specifically, the system acquires the pulse characteristic parameters of the load using a high-frequency data acquisition device to perceive the dynamic operating characteristics of the high-frequency pulse load in real time. Based on these pulse characteristic parameters, an electrochemical relaxation time constant model is constructed using electrochemical kinetics principles to quantify the microscopic relaxation mechanism within the battery. Battery operating parameters are input into the model to calculate the battery's effective recovery window, which is then compared with the load's natural rest time to generate a time-domain matching result, used to evaluate the matching relationship between load intermittent periods and battery recovery capability. Responding to the matching result indicating insufficient natural rest time, the energy storage battery cluster is divided into multiple virtual subsets in the logic control layer, and their alternating discharge states are maintained, enabling dynamic grouping and rotational scheduling of the battery pack without increasing hardware. A multi-objective collaborative scheduling model is constructed with the objective functions of minimizing cumulative polarization voltage and maximizing effective discharge capacity, and an optimization algorithm is used to solve for the optimal control sequence, used to generate a collaborative control strategy that balances polarization suppression and capacity release. Based on the optimal control sequence, the power converter executes power allocation and state switching operations for each virtual subset, used to complete precise closed-loop control.

[0024] In the overall technical solution, this invention addresses the lack of microscopic mechanisms in the control model described in the background technology. By constructing a relaxation time constant model based on electrochemical kinetics and calculating the effective recovery window, the microscopic recovery effect of the battery is transformed into a quantifiable time parameter, thus providing key timing basis for control decisions. This solves the defect of existing technologies that cannot actively utilize the recovery effect to improve performance due to neglecting electrochemical relaxation characteristics. Regarding the high cost caused by hardware dependence, this invention dynamically divides and maintains multiple virtual subsets in the logic control layer based on time-domain matching results, realizing time-sharing multiplexing and rotation control of the battery pack in a software-defined manner. This eliminates the need for additional hardware such as supercapacitors or simple capacity over-provisioning, solving the drawbacks of high life-cycle costs and complex topologies caused by hardware stacking in traditional solutions. Finally, addressing the problem of polarization accumulation restricting capacity utilization, this invention constructs a multi-objective collaborative scheduling model with minimizing accumulated polarization voltage as one of its core objectives. By directly associating the state switching of each virtual subset during control execution, it achieves active management and suppression of polarization voltage, solving the technical bottleneck of "pseudo-empty battery" phenomenon caused by continuous high-rate discharge and the resulting low effective capacity release rate. Therefore, the technical solution of the present invention solves the technical problem that the existing electrochemical energy storage life management technology lacks adaptability of control strategies under high-frequency pulse conditions, and improves the dynamic response adaptability, operating economy and battery capacity release efficiency of the energy storage system under pulse load scenarios. Attached Figure Description

[0025] Figure 1 This is a flowchart of an embodiment of the energy storage pulse collaborative control method based on electrochemical recovery according to the present invention;

[0026] Figure 2 This diagram illustrates the process of dynamically acquiring pulse characteristic parameters using a high-frequency data acquisition device according to another embodiment of the present invention.

[0027] Figure 3 This diagram illustrates the process of calculating battery state parameters and quantizing the effective recovery window using the electrochemical relaxation time constant model according to another embodiment of the present invention.

[0028] Figure 4 This diagram illustrates the logic control layer execution time matching coefficient calculation and grouping strategy decision-making flowchart according to another embodiment of the present invention.

[0029] Figure 5 This diagram illustrates the process of solving the predictive control optimization problem using a multi-objective cooperative scheduling model according to another embodiment of the present invention.

[0030] Figure 6 This is a schematic diagram of the structure of an energy storage pulse collaborative control system based on electrochemical recovery according to an embodiment of the present invention;

[0031] Figure 7 This is a block diagram of an electronic device used to implement embodiments of the present invention. Detailed Implementation

[0032] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of the present invention, including various details to aid understanding. These details should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of the invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0033] During the development of this invention, the inventors, through extensive experiments and data analysis, revealed the intrinsic relationship between load pulse characteristic parameters (such as natural rest time) and battery electrochemical relaxation characteristics (such as effective recovery window): the intermittent duty cycle characteristics of the load and the ion diffusion relaxation time constant inside the battery have a time-domain coupling relationship. That is, after high-rate discharge, the battery needs a specific resting time to eliminate polarization effects, and if the natural rest time of the load is insufficient, polarization will accumulate. Based on this relationship, the inventors innovatively proposed this technical solution, which uses a high-frequency data acquisition device to acquire the dynamic operating conditions of the load in real time, quantifies the battery recovery requirements by constructing an electrochemical relaxation time constant model, and combines a dynamic virtual grouping strategy to realize time-division multiplexing of battery clusters. This achieves active suppression of polarization, improvement of effective capacity release rate, and extension of battery cycle life under pulse conditions, reflecting the core concept of the "software-defined physical layer" active breathing and recovery mode.

[0034] Specifically, through comparative experiments, the invention team discovered that traditional continuous discharge or simple capacity oversizing methods suffer from technical defects that ignore the microscopic electrochemical relaxation characteristics of the battery: relying solely on macroscopic state of charge (SOC) for linear scheduling fails to cope with the accumulation of polarization voltage caused by high-frequency pulses. These technical defects lead to premature battery voltage cutoff (pseudo-empty battery), low effective capacity utilization, and shortened cycle life. However, the proposed collaborative control method based on electrochemical recovery mechanisms can improve energy release efficiency under pulsed conditions. Through a dynamic virtual grouping and rotation strategy, orderly switching between pressure-bearing discharge and forced relaxation states of battery cells can be achieved. By matching load intervals and battery recovery windows in the time domain, sufficient depolarization time can be ensured for each virtual subset. Through multi-objective optimization scheduling aimed at minimizing accumulated polarization voltage, collaborative optimization of electrochemical performance throughout the entire cycle can be achieved.

[0035] Therefore, this invention provides a pulse-based coordinated control method for energy storage based on electrochemical recovery, applicable to user-side electrochemical energy storage systems (hereinafter referred to as "the system"). This system can be deployed locally or collaboratively in the cloud on the energy management platform of the energy storage system to achieve battery coordinated control and lifespan optimization under high-frequency pulse load conditions. Specifically, this system can be deployed in various hardware environments, including but not limited to: industrial control servers integrating high-frequency data acquisition devices, local controllers of energy storage converters with edge computing capabilities, and cloud computing platforms supporting model prediction algorithms. This flexible deployment architecture allows the system to meet the high requirements for real-time control and reliability of centralized energy storage power stations in large industrial and commercial parks, while also adapting to the lightweight deployment needs of distributed, modular energy storage units in resource-constrained scenarios.

[0036] like Figure 1 As shown, the method may include:

[0037] Step S110: Obtain the pulse characteristic parameters of the load through a high-frequency data acquisition device.

[0038] Among them, pulse characteristic parameters refer to the dynamic operating condition characteristics extracted by real-time monitoring of the load power waveform, including pulse peak power, single pulse duration (…). ) and the natural pause time of the pulse interval ( The peak power of a pulse represents the maximum power value reached by the load during the pulse, the duration of a single pulse represents the length of time the pulse lasts from start to end, and the natural pause time represents the interval between adjacent pulses.

[0039] Specifically, the system can be achieved by deploying high-frequency data acquisition devices (such as those with a sampling frequency of 10 kHz). The Hall sensor collects the load current ripple in real time and uses a sliding time window algorithm (window duration is 60 seconds). The step size is 100 milliseconds. The power data is discretized and sampled to identify power abrupt change points and calculate the above parameters.

[0040] For example, when the system detects that the power value exceeds a preset threshold (such as 1.5 times the average base load) and the duration exceeds the de-jitter time limit, it marks the pulse start time. and end time Then calculate the duration of a single pulse. Natural rest time And extract the peak power within the window. .

[0041] In other embodiments, such as Figure 2The diagram shows the process flow of a high-frequency data acquisition device dynamically acquiring pulse characteristic parameters. It details the complete steps the system takes within one working cycle, from real-time data acquisition to the calculation and storage of characteristic parameters. The process begins with a "start" command, followed by sampling at a frequency of approximately 10 kHz. The sensor performs a "high-frequency data acquisition" operation, converting the continuous load current signal into a high-fidelity discrete-time sequence. Then, the system performs multiplication operations on the raw current and voltage data to generate a discretized "power sequence". To focus on the current operating conditions, the system adopts a window duration (…). The "sliding window truncation" method, with a time limit of 60 seconds, extracts time-sensitive analysis segments from long sequences. Subsequently, the process proceeds to the crucial "threshold determination" step. In the "step", the power data within the window is compared with the preset power threshold. The comparison is performed; if the result is "No", the system returns to the "Continue Monitoring" state and waits for the next data window; if the result is "Yes", the pulse event is confirmed, and the pulse start is identified sequentially. "and "end of monitoring pulse" Based on accurate timestamps, the system "calculates characteristic parameters," specifically including pulse peak power (…). ), duration of a single pulse ( ) and natural rest time ( Furthermore, the system "applies the following formula." Calculate duty cycle This process quantifies the pulse duty cycle of the load. Ultimately, all characteristic parameters are "stored in a dynamic fingerprint database," providing real-time data support for subsequent model building and marking the "end" of the current cycle or the start of the "next cycle." This implementation clearly illustrates how the present invention achieves automated and precise sensing of load pulse characteristics through programmed steps, ensuring the timeliness and reliability of pulse characteristic parameters and laying a solid data foundation for subsequent recovery control.

[0042] Step S120: Construct an electrochemical relaxation time constant model based on pulse characteristic parameters and electrochemical kinetics principles.

[0043] The electrochemical relaxation time constant model is a mathematical model that describes the polarization recovery process by quantifying the ion diffusion kinetics within the battery. Its core output is the electrochemical relaxation time constant. It is used to characterize the time scale required for the battery terminal voltage to recover to a steady state.

[0044] Specifically, the system can be constructed by building a second-order resistive-capacitive structure for the battery. Equivalent circuit model (including ohmic internal resistance, electrochemical polarization) Loop and concentration polarization (loop), and based on Fick Diffusion law analyzes the diffusion process of lithium ions in the solid phase particles of the electrode, and the relaxation time constant is derived by combining battery operating parameters (such as state of charge, temperature T, and aging degree SOH). SOH stands for State of Health. In the field of electrochemical energy storage, this term is used to quantify the percentage of a battery's current performance (such as capacity and internal resistance) relative to its initial performance, and is a key parameter characterizing the degree of battery aging and remaining lifespan.

[0045] For example, the system uses a formula Calculate the concentration relaxation constant, where Electrode particle radius (unit: micrometers) ), Solid-phase diffusion coefficient (unit: square centimeters per second) ), It is an aging correction factor.

[0046] Step S130: Input the battery operating parameters into the electrochemical relaxation time constant model, calculate the effective recovery window of the battery through the electrochemical relaxation time constant model, and compare the natural pause time in the pulse characteristic parameters with the effective recovery window to generate time-domain matching results.

[0047] The effective recovery window refers to the shortest resting time required for the battery's back-end voltage to recover to a specific percentage (e.g., 95%) of its steady-state value after high-rate discharge. The temporal matching result is obtained by comparing natural pause times. With effective recovery window Decision indicators generated from the size relationship, such as the matching coefficient. As a basis for judgment.

[0048] Specifically, the system can calculate Meet the conditions Time required and will With threshold (e.g.) Comparison: If If the natural intervals are insufficient, a grouping strategy needs to be triggered. For example, when... Second, seconds, The system generates a matching result for "intermittent insufficiency".

[0049] In other embodiments, such as Figure 3The diagram shows the process of calculating battery state parameters and quantizing the effective recovery window using the electrochemical relaxation time constant model. It details the process by which the model calculates key time constraints directly applicable to control decisions based on real-time battery state parameters. The process begins with the "Start" command, first constructing a second-order RC equivalent circuit model of the battery based on the current operating conditions. This model explicitly includes the ohmic internal resistance (…). ), electrochemical polarization circuit ( ) and concentration polarization circuit ( Next, the process enters the "based on Fick ( The second law analyzes the solid-state diffusion process, and the rate-determining step in the process is determined by solving the diffusion equation. Subsequently, the system acquires real-time battery state parameters, including state of charge (SOC, unitless), operating temperature (T, unit is degrees Celsius °C) and aging degree (SOH, unitless). Based on the above parameters, the system applies formula (1) to calculate the decay relaxation time constant. The formula is ,in Electrode particle radius (unit: micrometer) ), Represents the solid-phase diffusion coefficient (unit: square centimeters per second). ), This is an effective aging factor. Then, the process sets an "effective recovery threshold condition" based on technical requirements, for example, aiming to restore the polarization voltage to below 5% of its initial value. Furthermore, the system calculates the effective recovery window. This means finding the shortest settling time required to satisfy the aforementioned threshold conditions. Ultimately, the model outputs the macroscopic time constraint variable. (Unit: seconds), this variable is the core output "effective recovery window" that can be directly used to match the time domain, and the process then "ends". This implementation clearly illustrates how the present invention combines the microscopic electrochemical parameters of the battery (such as diffusion coefficient, particle radius) with the macroscopic operating state (SOC, temperature, aging), and derives accurate and operable time constraints through physical models, providing a key quantitative and decision-making basis for the entire collaborative control strategy.

[0050] Step S140: In response to the time-domain matching result indicating that the natural pause time is less than the effective recovery window, the energy storage battery cluster is divided into multiple virtual subsets in the logic control layer of the energy management system, and at least one virtual subset is in a discharge state at any given time.

[0051] Among them, virtual subsets refer to several independent groups that are dynamically divided into physically connected battery clusters at the logic layer through software definition; maintaining discharge state refers to ensuring continuous power output by controlling the alternating switching of each subset.

[0052] Specifically, the system can calculate the minimum number of groups. (in (Indicates rounding up), and sets the phase difference. This enables the subset to work in a time-division multiplexing mechanism. For example, when... Second, Second, seconds, The system divides the battery cluster into two subsets with a phase difference of π radians (rad) to ensure that at any given time, one subset is discharging.

[0053] In other embodiments, such as Figure 4 The diagram shows the flowchart of the logic control layer's execution time matching coefficient calculation and grouping strategy decision-making. It specifically illustrates how the system automatically selects the control mode and generates grouping instructions based on the time-domain parameters of the load and battery through quantified matching relationships. The process begins with the "Start" instruction, first executing the "Input Parameters" command. "Steps, in which" The duration of a single pulse. For natural rest time, This constitutes one complete pulse cycle. The process then proceeds to "calculate the time matching coefficient". The step involves a coefficient that quantifies the proportion of the load pulse duration within the entire cycle. This is followed by the critical decision node, "Judgment." This judgment is equivalent to assessing whether the duration of a single pulse is equal to or exceeds a complete cycle, which indicates whether the load is a continuous pulse or an extreme condition (theoretically). (Settings for model fault tolerance or special continuous pulses). If the judgment result is "yes" (i.e.) If the result is "No" (i.e., ...), the process enters the "Mode B: Trigger Inter-group Grouping Strategy" path, indicating that the load duty cycle is extremely high and virtual subset partitioning must be forcibly initiated; If the process enters "Mode A: Full-flow synchronous operation", then the process will enter "Mode A: Full-flow synchronous operation". "Path B" means that no virtual subsets are divided, and the entire battery cluster responds to the load synchronously. Under path B, the subsequent process will "apply the formula..." And the derivation formula for calculating the minimum number of groups The calculation is based on , With effective recovery window Once the parameters are finalized, the minimum number of subgroups required to meet battery recovery needs is determined. Then, the system logically divides each subgroup accordingly. The system logically divides battery cluster resources into blocks based on capacity or power ratio. Then, it calculates the input phase difference. This involves planning precise timing intervals for each virtual subset. Finally, all decision results are aggregated in the "output grouping instructions and phase parameters" step, generating a control instruction package containing information such as switching status, number of groups, and phase difference, at which point the process "ends." This implementation clearly illustrates how the present invention transforms the macroscopic duty cycle characteristics of the load into decision variables. Based on this variable, intelligent switching of control modes and precise parameter generation are achieved, ensuring the adaptability of the control strategy to complex dynamic working conditions.

[0054] Step S150: Construct a multi-objective cooperative scheduling model with the objective functions of minimizing the cumulative active voltage and maximizing the effective discharge capacity, and use an optimization algorithm to solve for the optimal control sequence of the multi-objective cooperative scheduling model.

[0055] The objective function refers to the composite index in the mathematical expression that simultaneously optimizes polarization voltage suppression and capacity release; the optimal control sequence includes the switching times and power allocation ratio sequences of each virtual subset.

[0056] Specifically, the system can construct the objective function using a Model Predictive Control (MPC) algorithm to solve multi-objective optimization problems in energy storage pulse coordinated scheduling, achieving dynamic prediction and optimal control of battery state. Furthermore, constraints such as terminal voltage and temperature rise rate are applied, and a control sequence is generated by solving a quadratic programming problem. For example, in the prediction time domain... Within seconds, the system uses weighting coefficients. Optimize and output subset power allocation ratio (like ).

[0057] In other embodiments, such as Figure 5 The diagram shows the optimization solution flowchart for Model Predictive Control (MPC) in a multi-objective cooperative scheduling model. It illustrates how the system, at each control time step, uses a rolling time-domain optimization algorithm to transform the complex multi-objective programming problem into a sequence of control commands that can be solved online, thereby achieving closed-loop dynamic adjustment of the virtual subset's state and power allocation. The process begins with the "Start / Next Time k" command, first executing "Get Input (Group A, Load Forecast)". Battery status The "load prediction" step, where "Group A" represents the logical identifier and status information of all current virtual subsets, is called "load prediction". "Refers to the estimated load power demand over a future period based on real-time pulse characteristic parameters (unit: kilowatts)" "Battery status" "Refers to the real-time state of charge (unitless) and temperature (unit: degrees Celsius) of each virtual subset battery." Subsequently, the system executes the "MPC controller initialization (setting time domain T)" step, setting the total prediction step size N for this rolling optimization (e.g., ...). Step, control step length If the time domain is specified in seconds, then the prediction time domain is determined. (seconds). Next, the process enters the core area of ​​the "rolling optimization loop" marked with a dashed box. Within this loop, the processes sequentially execute: "updating the state-space equations and observers," which dynamically corrects the internal state of the prediction model based on the latest battery and load data; and "constructing the multi-objective optimization function." (formula ")," which defines a mathematical objective function that includes a cumulative positive voltage minimization term and an effective discharge capacity maximization term; executes "introducing safety constraints (voltage / temperature rise / power balance)," which applies constraints on the battery terminal voltage range, individual cell temperature rise rate, and system power balance equations; executes "solving the optimization problem (e.g., QP)." The solver is a numerical computation kernel that generates values ​​from the current time step. To the future "Optimal control sequence" at time 1 Subsequently, the system extracts only the first instruction from the sequence and executes "output current moment control instruction," which contains the specific switching state and power allocation ratio for each virtual subset. After output, the process enters the "waiting for the next sample cycle" state, completing the control action for the current moment. When the next control cycle begins, the system will repeat the above "start / next moment" process based on the latest feedback data. This process forms a closed loop. This implementation clearly illustrates how the present invention closely combines "dry experiment" (model prediction optimization) with "wet experiment" (real-time feedback and execution). Through a rolling optimization mechanism, the multi-objective collaborative scheduling model can adapt to changes in system state in real time, continuously generate optimal control strategies that take into account performance, safety and efficiency, and ultimately achieve intelligent and high-precision collaborative operation of the energy storage system under pulse load conditions.

[0058] Step S160: Based on the optimal control sequence, the power allocation and state switching operations of each virtual subset are executed through the power converter to complete the energy storage pulse coordinated control.

[0059] Among them, power converter refers to energy storage converter (Power Conversion System, abbreviated as PCS, in the field of electrochemical energy storage, this term specifically refers to the core power electronic equipment that realizes bidirectional conversion between battery DC power and AC grid power) or multi-channel DC-DC controller; state switching operation refers to the switching of the control subset between the "pressure discharge" and "forced relaxation" states.

[0060] Specifically, the system can control automation technology via a high-speed communication bus (such as Ethernet). A deterministic and real-time-enabled industrial fieldbus communication protocol based on Ethernet sends control sequences to the power converter to adjust the power shunt ratio of each subset in real time. The terminal voltage recovery curve is monitored to evaluate the depolarization effect.

[0061] For example, after the system performs the switch, the discharge cutoff time is reduced from 45.2 minutes compared to the traditional strategy. Extended to 58.5 minutes, total energy released increased from 185.6 kWh. Increased to 238.3 kWh, energy efficiency improvement rate .

[0062] Therefore, according to the above implementation method, the system is achieved through six core steps: pulse characteristic parameter acquisition, electrochemical relaxation time constant model construction, time domain matching result generation, logical virtual subset partitioning, multi-objective cooperative scheduling model solving, and power allocation execution. Specifically, the system acquires the pulse characteristic parameters of the load through a high-frequency data acquisition device to perceive the dynamic operating characteristics of the high-frequency pulse load in real time. Based on the pulse characteristic parameters, an electrochemical relaxation time constant model is constructed using electrochemical kinetics principles to quantify the microscopic relaxation mechanism inside the battery. Battery operating parameters are input into the model to calculate the battery's effective recovery window, which is then compared with the natural rest time of the load to generate a time-domain matching result, used to evaluate the matching relationship between load intermittent and battery recovery capability. In response to the matching result indicating insufficient natural rest time, the energy storage battery cluster is divided into multiple virtual subsets in the logic control layer, and their alternating discharge states are maintained, enabling dynamic grouping and rotation scheduling of the battery pack without increasing hardware. A multi-objective cooperative scheduling model is constructed with the objective functions of minimizing cumulative polarization voltage and maximizing effective discharge capacity, and an optimization algorithm is used to solve for the optimal control sequence, used to generate a cooperative control strategy that takes into account both polarization suppression and capacity release. According to the optimal control sequence, the power converter executes the power allocation and state switching operations of each virtual subset to complete precise closed-loop control.

[0063] Specifically, in this implementation, addressing the lack of microscopic mechanisms in the control model mentioned in the background technology, a relaxation time constant model based on electrochemical kinetics is constructed and the effective recovery window is calculated. This transforms the battery's microscopic recovery effect into a quantifiable time parameter, providing crucial timing information for control decisions. This solves the problem of existing technologies failing to actively utilize the recovery effect to improve performance due to neglecting electrochemical relaxation characteristics. Regarding the high cost caused by hardware dependence, multiple virtual subsets are dynamically partitioned and maintained in the logic control layer based on time-domain matching results. This software-defined approach enables time-sharing reuse and rotation control of the battery pack, eliminating the need for additional hardware such as supercapacitors or simple capacity over-provisioning. This solves the drawbacks of high life-cycle costs and complex topologies caused by hardware stacking in traditional solutions. Finally, addressing the problem of polarization accumulation restricting capacity utilization, a multi-objective collaborative scheduling model with minimizing accumulated polarization voltage as one of its core objectives is constructed. The state switching of each virtual subset is directly correlated during control execution, achieving active management and suppression of polarization voltage. This solves the technical bottleneck of the "pseudo-empty battery" phenomenon caused by continuous high-rate discharge and the resulting low effective capacity release rate. Therefore, the technical solution of the present invention solves the technical problem that the existing electrochemical energy storage life management technology lacks adaptability of control strategies under high-frequency pulse conditions, and improves the dynamic response adaptability, operating economy and battery capacity release efficiency of the energy storage system under pulse load scenarios.

[0064] In other embodiments, Table 1 below shows specific calculation examples of the dynamic virtual grouping strategy of energy storage units using the method proposed in this invention under three typical high-frequency pulse load conditions. As shown in Table 1, by obtaining the pulse duration of the operating condition ( ), load interval time ( ) and the battery's necessary recovery window calculated based on the electrochemical relaxation model ( The time-domain matching coefficient can be calculated. .according to The system automatically determines the value and generates corresponding control strategies: under the operating conditions Down, The system is judged to have "sufficient intermittent time" and adopts Mode A (minimum number of groups). This means that a single virtual subset continuously handles the pulse load; under operating conditions Down, The system was determined to be "intermittently insufficient" and switched to the appropriate mode. Calculations show that at least two virtual subsets are required. To work alternately and set the input phase difference (i.e., 180), ensuring that one group is working while the other is in recovery; under working conditions Down, If it is determined to be "severely insufficient", the system will also adopt the same mode. However, calculations show that at least four virtual subsets are required. Perform refined time-sharing multiplexing and set up (i.e., 90) to achieve denser power relay. This example clearly demonstrates that the present invention can dynamically and quantitatively determine the optimal number of groups and the coordinated phase based on real-time time-domain matching results, thereby ensuring that the battery recovery time requirements are met under any pulse characteristics, and achieving synergy between life protection and power support.

[0065]

[0066] In other embodiments, Table 2 below shows the performance comparison data of the proposed cooperative control strategy and the traditional continuous discharge strategy under a typical periodic high-frequency pulse load condition (pulse width 2 minutes, interval 1 minute). As can be seen from Table 2, the strategy of the present invention achieves a significant improvement in overall performance through dynamic virtual grouping and phase-coordinated scheduling. Compared with the traditional strategy, the strategy of the present invention extends the effective discharge cutoff time of the system from 45.2 minutes to 58.5 minutes, and increases the total released energy from 185.6 kWh to 238.3 kWh. This is directly due to the alternating grouping working mode fully guaranteeing the battery relaxation recovery time and slowing down the accumulation of polarization voltage. Simultaneously, the peak temperature rise rate is reduced from... Significantly reduced to This demonstrates that the time-sharing multiplexing mechanism of the present invention effectively distributes the heat load and alleviates the transient heat generation pressure on the battery. Ultimately, the strategy of the present invention achieves a 28.4% energy efficiency improvement under this operating condition. This example, through a quantitative comparison of key performance indicators, intuitively verifies the substantial progress of the present invention in extending the continuous power supply time of the system, increasing the total output energy, improving thermal safety, and enhancing overall energy efficiency.

[0067]

[0068] In some embodiments, the high-frequency data acquisition device is configured with a current monitoring unit, a power sampling unit, and a feature extraction unit; acquiring the pulse characteristic parameters of the load through the high-frequency data acquisition device includes:

[0069] Real-time monitoring data of load current ripple is obtained through the current monitoring unit.

[0070] Among them, the real-time monitoring data of load current ripple refers to the AC fluctuation component superimposed on the load current waveform obtained by high-frequency sampling (e.g., 10 kHz), which is used to characterize the instantaneous dynamic characteristics of the load.

[0071] Specifically, the current monitoring unit uses a Hall effect sensor or a Rogowski coil to continuously acquire load current signals at a high sampling rate, and extracts ripple components through digital filtering algorithms (such as a low-pass filter) to form time-series data. For example, during the lifting operation of a port crane, the system detects a sudden increase in the load current ripple amplitude from a steady-state 5 amps to a peak value of 25 amps, with a ripple frequency of 100 Hz, thereby identifying the start of the pulse event.

[0072] The load power is discretized by using a sliding time window algorithm through a power sampling unit to obtain load power sampling data.

[0073] Among them, the sliding time window algorithm refers to a data segmentation method based on a fixed-duration window sliding along the time axis, which is used to discretize continuous power signals; the load power sampling data refers to the discrete sequence of power values ​​within each window.

[0074] Specifically, the power sampling unit calculates the instantaneous power in real time with a window duration of 60 seconds and a sliding step size of 100 milliseconds. This generates power-time series data. For example, the system at time point... Instant Within a 100-second window, 600 power data points were sampled (intervals of 100 milliseconds), with a peak power of 50 kilowatts. This is used for subsequent analysis.

[0075] The feature extraction unit identifies abrupt changes in load power based on load power sampling data. These abrupt changes include the pulse start time and the pulse end time.

[0076] Among them, the mutation point refers to a significant change point in the load power sequence that exceeds a preset threshold (such as 1.5 times the average value of the base load) and lasts for a duration greater than the de-jittering time limit (such as 10 milliseconds); the pulse start time indicates the starting timestamp of the power jump from the baseline, and the pulse end time indicates the ending timestamp of the power falling back to the baseline.

[0077] Specifically, the feature extraction unit uses a differential thresholding method to calculate the first-order difference value of the power sequence. When the difference value continuously exceeds the threshold, abrupt change points are marked, and the start and end times are recorded using timestamps. For example, when the power jumps from 5 kW to 45 kW and lasts for 20 seconds, the system marks the start time. Seconds, end time Second.

[0078] The duration of a single pulse is calculated based on the pulse start time and pulse end time.

[0079] The duration of a single pulse refers to the time interval from the start of the pulse to the end of the pulse, and is used to quantify the duration of a pulse event.

[0080] Specifically, the system calculates the duration of a single pulse using the time difference. The result is in seconds. Units. For example, if Second, seconds, then Second.

[0081] The natural pause time is calculated based on the pulse start and pulse end times of adjacent pulses.

[0082] The natural pause time refers to the interval between adjacent pulse events when the load is in the baseline state, that is, the time difference between the end of the previous pulse and the start of the next pulse.

[0083] Specifically, the system extracts the timestamps of continuous pulse sequences and calculates the natural pause time. ,in This indicates the pulse sequence number. For example, if the first pulse ends at... Seconds, the start time of the second pulse seconds, then Second.

[0084] The peak power of the pulse is extracted based on the load power sampling data.

[0085] Among them, peak pulse power refers to the maximum value of the load power reached during a single pulse event.

[0086] Specifically, the feature extraction unit searches for the maximum value of the power sequence within the identified pulse time window, i.e. For example, if the maximum power value in the sampled data within the pulse window is 50 kilowatts, then... kilowatt.

[0087] Pulse characteristic parameters are generated based on the duration of a single pulse, the natural pause time, and the peak power of the pulse.

[0088] Among them, the pulse characteristic parameters refer to a set of characteristic vectors used to describe the load pulse condition, including the duration of a single pulse. Natural rest time and peak pulse power .

[0089] Specifically, the system encapsulates the three parameters into a data structure (such as JSON format or an array) and adds a timestamp identifier for subsequent model input. For example, the system generates a parameter set. , as input to the electrochemical relaxation time constant model.

[0090] Therefore, according to the above implementation method, the system can capture the dynamic pulse characteristics of the load in real time and accurately, providing a reliable data basis for subsequent electrochemical recovery control, and avoiding the control lag problem caused by insufficient sampling frequency or coarse feature extraction in traditional methods.

[0091] In some embodiments, an electrochemical relaxation time constant model is constructed based on pulse characteristic parameters and combined with electrochemical kinetic principles, including:

[0092] Based on the pulse peak power and single pulse duration in the pulse characteristic parameters, combined with the battery operating parameters, a second-order resistive-capacitive equivalent circuit model of the battery is constructed. The second-order resistive-capacitive equivalent circuit model includes ohmic internal resistance, a first-order resistive-capacitive circuit characterizing electrochemical polarization, and a second-order resistive-capacitive circuit characterizing concentration polarization.

[0093] The second-order resistive-capacitive equivalent circuit model refers to an electrical network model used to simulate the dynamic response characteristics inside a battery. The ohmic internal resistance characterizes the instantaneous voltage drop of the battery. The first-order resistive-capacitive circuit is used to describe the rapid electrochemical polarization process at the electrode interface, and the second-order resistive-capacitive circuit is used to describe the slow concentration polarization process caused by the diffusion of lithium ions inside solid particles.

[0094] Specifically, the system determines the current excitation range that the model needs to cover based on the peak power of the pulse, determines the time constant range of the model based on the duration of a single pulse, and dynamically corrects the model component parameters by combining the state of charge, temperature and aging degree in the battery operating parameters.

[0095] For example, when processing peak power of 50 kilowatts Duration is 30 seconds Under pulsed operating conditions, the equivalent circuit model of the system has an internal resistance of 2.5 milliohms. The time constant of the first-order resistor-capacitor circuit is set to 0.5 seconds to 2 seconds, and the time constant of the second-order resistor-capacitor circuit is set to 50 seconds to 200 seconds, in order to cover the polarization dynamic process from the second level to the hundred-second level.

[0096] Based on Fick's diffusion law, the diffusion process of lithium ions in the solid phase particles of the electrode is analyzed.

[0097] Fick's law of diffusion describes the physical law of diffusion transport of matter driven by a concentration gradient; the diffusion process specifically refers to the mass transport behavior of lithium ions within the solid phase particles of electrode active materials due to uneven concentration.

[0098] Specifically, the system calculates the variation of lithium ion concentration distribution in the radial direction of particles over time by solving the partial differential equation of Fick's second law in spherical coordinates, thereby quantifying the diffusion kinetics characteristics.

[0099] For example, for particles with a diameter of 5 micrometers ( The lithium iron phosphate cathode material was used. The system was set to an initial uniform concentration and constant current discharge conditions were applied at the boundary. The concentration difference between the particle center and the surface was obtained by numerical solution as a function of time. This curve reflects the relaxation characteristics of concentration polarization.

[0100] Based on the diffusion process and combined with battery operating parameters, the electrochemical relaxation time constant of the battery is derived to construct an electrochemical relaxation time constant model.

[0101] Among them, the electrochemical relaxation time constant refers to the characteristic time parameter that characterizes the speed of the battery polarization voltage recovery process; the electrochemical relaxation time constant model refers to the mathematical model that uses this time constant as the core output and can predict the battery voltage recovery behavior.

[0102] Specifically, the system couples the concentration relaxation time obtained from the diffusion process analysis with the electrochemical polarization relaxation time, and performs a weighted correction on the coupling time constant based on the state of charge in the battery operating parameters, and finally derives a comprehensive relaxation time constant applicable to the current operating conditions.

[0103] For example, the system calculates the concentration diffusion relaxation time constant to be 120 seconds and the electrochemical polarization relaxation time constant to be 1.5 seconds. When the state of charge is 60%, the comprehensive relaxation time constant is 95 seconds obtained through the weighting formula. This constant will be used as the direct input for the calculation of the effective recovery window.

[0104] Therefore, according to the above implementation method, the system can establish an electrochemical relaxation time constant model that accurately reflects the coupling relationship between the internal ion diffusion dynamics of the battery and the external pulse load characteristics, providing a key time scale benchmark for subsequent time-domain matching and coordinated control.

[0105] In some embodiments, the step of calculating the effective recovery window of the battery using an electrochemical relaxation time constant model and comparing the natural pause time in the pulse characteristic parameters with the effective recovery window to generate a time-domain matching result includes:

[0106] The effective recovery window value of the battery is calculated using a recovery window calculation algorithm configured with an electrochemical relaxation time constant model.

[0107] The effective recovery window refers to the time required for the battery terminal voltage to recover from the polarization state after high-rate discharge to a specific proportion of the thermodynamic equilibrium electromotive force, expressed in seconds. Units.

[0108] Specifically, the recovery window calculation algorithm calculates the electrochemical relaxation time constant by... The effective recovery window is calculated by multiplying by a natural logarithm coefficient corresponding to the target recovery ratio. The formula is as follows: ,in Indicates the target recovery percentage (e.g.) This indicates a recovery to 95% of the steady-state value. For example, when the electrochemical relaxation time constant... Seconds, target recovery rate At that time, the effective recovery window value is calculated. Second.

[0109] Compare the natural pause time with the effective recovery window value to generate a time-domain matching result.

[0110] Among them, the time-domain matching result refers to the comparison of the natural downtime of the load through quantification. ) and battery effective recovery window ( The decision identifier is generated based on the degree of matching between the two criteria.

[0111] Specifically, the system calculates the time-domain matching coefficients. and will With preset threshold (e.g.) Compare: If If so, a "sufficient match" result will be generated; if If so, an "insufficient match" result will be generated. For example, when Second, seconds, The system generates a time-domain matching result indicating "insufficient matching," triggering subsequent virtual grouping strategies.

[0112] The recovery window calculation algorithm is configured to calculate the shortest resting time required for the battery terminal voltage to recover to a specific proportion of the steady-state value defined by the electrochemical relaxation time constant model, based on the calculation of the electrochemical relaxation time constant.

[0113] Specifically, the algorithm transforms the electrochemical relaxation time constant into an operable time parameter with clear physical meaning by solving for the time required for the battery polarization voltage to decay to below 5% of its initial value.

[0114] For example, the algorithm sets the voltage recovery target as Combining the first-order RC circuit response model Inverse solution time get That is, when seconds, Second.

[0115] Therefore, according to the above implementation method, the system can transform the microscopic relaxation characteristics inside the battery into macroscopically usable time constraint parameters, and generate clear control decision basis through precise numerical comparison, providing key input for subsequent dynamic grouping and coordinated scheduling.

[0116] In some embodiments, the step of dividing the energy storage battery cluster into multiple virtual subsets in the logic control layer of the energy management system and maintaining that at least one virtual subset is in a discharged state at any given time includes:

[0117] The minimum number of groups is calculated based on the natural pause time, the effective recovery window, and the duration of a single pulse.

[0118] The minimum number of groups refers to the minimum number of virtual subsets that meet the requirements of the battery's effective recovery window, which is calculated through temporal matching relationships.

[0119] Specifically, the system uses formulas Calculate the minimum number of groups, where Indicates the natural pause time (unit: seconds) ), Indicates the effective recovery window (unit: seconds) ), Indicates the duration of a single pulse (unit: seconds). ),symbol This indicates the rounding up operation.

[0120] For example, when Second, Second, In seconds, the calculation is obtained That is, the minimum number of groups is 2.

[0121] Based on the minimum number of groups, the energy storage battery cluster is divided into multiple virtual subsets.

[0122] Virtual subsets refer to battery cell groups dynamically divided at the logic layer through software definition, which are physically connected but independently controlled.

[0123] Specifically, the system calculates the minimum number of groups. The energy storage battery clusters are divided equally or proportionally by capacity. Each virtual subset has its own independent switching control interface. For example, when At that time, the system will have a total capacity of 400 kWh. The battery clusters are divided into two virtual subsets: subset 1 has a capacity of 200 kWh and subset 2 has a capacity of 200 kWh.

[0124] Set the phase difference between each virtual subset so that each virtual subset discharges alternately, in order to maintain that at least one virtual subset is in a discharging state at any given time.

[0125] The phase difference refers to the time interval between the discharge states of each virtual subset, expressed in radians (rad) or time units.

[0126] Specifically, the system uses formulas Calculate the phase difference, where N is the actual number of groups, ensuring that each subset works in rotation according to the time-division multiplexing mechanism. For example, when the number of groups... At that time, phase difference That is, the input time of subset 1 and subset 2 differs by half a cycle, thus achieving alternating discharge.

[0127] Therefore, according to the above implementation method, the system can dynamically adapt to the load pulse characteristics, optimize resource utilization by minimizing the number of groups, and ensure the continuity of power output by using phase difference control, thereby maintaining the efficient and stable operation of the battery pack under pulse conditions.

[0128] In some embodiments, a multi-objective cooperative scheduling model is constructed with the objective functions of minimizing the cumulative active voltage and maximizing the effective discharge capacity, and an optimization algorithm is used to solve for the optimal control sequence of the multi-objective cooperative scheduling model, including:

[0129] Define the objective function of the multi-objective cooperative scheduling model, which includes minimizing the cumulative active voltage and maximizing the effective discharge capacity.

[0130] Among them, minimizing the cumulative polarization voltage means minimizing the sum of squares of the polarization voltages of all virtual subsets of batteries in the prediction time domain; maximizing the effective discharge capacity means maximizing the total energy released by the system to the load in the same period.

[0131] Specifically, the system uses a weighted summation method to transform a multi-objective problem into a single-objective optimization problem, with the objective function being... The mathematical expression is: ;For example, For control vectors, and Weighting coefficients (e.g.) ), For the first The subset in Polarization voltage at time , for Load power at any given time To control the step size (e.g., 1 second). For example, the system sets the prediction time domain (e.g., 60 seconds). Seconds, control step size The priority of polarization suppression and energy release is balanced by adjusting the weighting coefficients.

[0132] Safety constraints are set for the multi-objective cooperative scheduling model, including battery terminal voltage constraints, temperature rise rate constraints, and power balance constraints.

[0133] Among them, battery terminal voltage constraint refers to limiting the terminal voltage of a single battery cell to within an allowable range (e.g., V to (Voltage); Temperature rise rate constraint refers to limiting the rate of temperature change of the battery cell to no more than a threshold (e.g., ... Celsius per minute ( The power balance constraint requires that the sum of the output power of all virtual subsets equals the load demand power.

[0134] Specifically, the system applies the following constraints in each control cycle:

[0135] Terminal voltage constraint: ;

[0136] Temperature rise constraint: ;

[0137] Power balance: ;

[0138] For example, when the load power At kilowatts (kW), the system adjusts It ensures power balance while monitoring whether the voltage and temperature rise of each subset exceed the limits.

[0139] An optimization algorithm is used to numerically solve a multi-objective cooperative scheduling model with objective functions and security constraints, generating the optimal control sequence.

[0140] The optimization algorithm adopts the Model Predictive Control (MPC) framework, which transforms the problem into a Quadratic Programming (QP) problem through rolling time-domain optimization.

[0141] Specifically, at each control moment, the system initializes the QP problem based on the current state and calls a numerical solver (such as the interior-point method or the effective set method) to calculate the future... The optimal control sequence is obtained by executing only the first step of the control input and then rolling to the next cycle to solve it again.

[0142] For example, the system in The control sequence is obtained by solving the second. ,implement Afterwards, The sequence is updated again based on the latest state every second.

[0143] The optimal control sequence includes the switching state sequence and power allocation ratio sequence of each virtual subset.

[0144] Specifically, the switching state sequence refers to the switching state of each virtual subset at each control step in the prediction time domain (0 represents relaxation, 1 represents discharge); the power allocation ratio sequence refers to the power shunting ratio of each subset in the discharge state. The switching state sequence is a binary matrix. The power allocation ratio sequence is a continuous value matrix. And satisfy .

[0145] For example, for a scenario with two subsets and a three-step prediction time domain, the generated sequence is:

[0146] Throwing and cutting status: ;

[0147] Power distribution: .

[0148] Indicates that subset 1 is in Discharges at all times, subset 2 in It discharges continuously, and the power is distributed proportionally.

[0149] Therefore, according to the above implementation method, the system can accurately coordinate polarization suppression and capacity release requirements through a multi-objective optimization model, generate dynamic optimal control instructions under strict safety constraints, and provide a quantitative basis for the round-robin scheduling of virtual subsets.

[0150] In some embodiments, according to the optimal control sequence, power allocation and state switching operations of each virtual subset are performed by the power converter to complete the energy storage pulse coordinated control, including:

[0151] The optimal control sequence is sent to the power converter via the communication bus.

[0152] The communication bus refers to a high-speed industrial fieldbus used to transmit control commands between the energy management system and the power converter; the power converter refers to an energy storage converter or a (DC-DC) converter used to perform power distribution and state switching operations.

[0153] Specifically, the system utilizes high-speed industrial fieldbuses (such as Ethernet control automation technology). The optimal control sequence is transmitted from the logic control layer of the energy management system to the local controller of the power converter in millisecond-level communication cycles. For example, the system configuration... The bus communication cycle is 1 millisecond. The optimal control sequence, which includes the switching state sequence and the power allocation ratio sequence, is encapsulated into a data frame and sent to the multi-channel DC-DC converter in real time.

[0154] The power converter controls each virtual subset to switch between the pressurized discharge state and the forced relaxation state based on the switching state sequence and the power allocation ratio sequence in the optimal control sequence.

[0155] Among them, the switching state sequence refers to the switching state sequence of each virtual subset at each control step in the prediction time domain (0 represents the forced relaxation state, and 1 represents the pressure discharge state); the power allocation ratio sequence refers to the power shunting ratio sequence of each virtual subset in the discharge state.

[0156] Specifically, the power converter parses the received sequence, controls the physical switching of each virtual subset according to the binary instructions of the switching state sequence, and adjusts the output current of the DC / DC converter according to the power allocation ratio sequence to achieve precise power allocation.

[0157] For example, for two virtual subsets, in controlling the step size At seconds, the switching state sequence is as follows: The power allocation ratio sequence is This indicates that subset 1 is in a pressurized discharge state and is allocated 80% of the load power, while subset 2 is in a forced relaxation state; At the specified time, the state switches to... The proportion sequence is This enables subset rotation.

[0158] Monitor the voltage recovery curve of each virtual subset.

[0159] Among them, the terminal voltage recovery curve refers to the transient characteristic curve of the terminal voltage recovering over time after the virtual subset switches from the pressure discharge state to the forced relaxation state, which is used to quantify the depolarization effect.

[0160] Specifically, the system uses a high-frequency voltage sensor (such as one with a sampling frequency of 10 kHz). Real-time acquisition of terminal voltage data of a subset in the relaxation state, plotting of voltage-time curves, and calculation of voltage rebound amplitude.

[0161] For example, during the relaxation phase, the terminal voltage of subset 1 drops from 3.0 volts. The voltage rebounded to 3.4 volts within 10 seconds, with a rebound amplitude of 0.4 volts, indicating that the polarization voltage was effectively eliminated.

[0162] In response to the detection of a sudden change in load characteristic parameters, a process of re-extracting pulse characteristic parameters and recombining virtual subsets is triggered.

[0163] Among them, a sudden change in load characteristic parameters refers to a significant change in key parameters such as the peak power of the load pulse, the duration of a single pulse, or the natural pause time (such as exceeding the preset tolerance band).

[0164] Specifically, the system compares the deviation of the current load characteristic parameters with the historical fingerprint database in real time. If the rate of change of any parameter exceeds the threshold (such as the rate of change of pulse peak power greater than 20%), the system will immediately trigger the re-extraction of pulse characteristic parameters and the reorganization of virtual subsets.

[0165] For example, when the peak power of the pulse is monitored to be 50 kilowatts When the load suddenly increases to 70 kilowatts and lasts for more than 5 seconds, the system determines that the load characteristics have changed abruptly and automatically restarts feature extraction and grouping strategy optimization.

[0166] Therefore, according to the above implementation method, the system can realize the dynamic rotation of virtual subsets through high-speed communication and precise control, and combined with real-time monitoring and adaptive reorganization mechanism, ensure the robustness and efficiency of energy storage pulse collaborative control.

[0167] Figure 6 This is a structural block diagram of an embodiment of the energy storage pulse collaborative control system based on electrochemical recovery according to the present invention.

[0168] like Figure 6 As shown, the energy storage pulse coordinated control system based on electrochemical recovery includes:

[0169] The pulse characteristic parameter acquisition module 210 is used to acquire the pulse characteristic parameters of the load through a high-frequency data acquisition device.

[0170] The time constant model construction module 220 is used to construct an electrochemical relaxation time constant model based on pulse characteristic parameters and combined with the principles of electrochemical kinetics.

[0171] The time-domain matching result generation module 230 is used to input battery operating parameters into the electrochemical relaxation time constant model, calculate the effective recovery window of the battery through the electrochemical relaxation time constant model, and compare the natural pause time in the pulse characteristic parameters with the effective recovery window to generate time-domain matching results.

[0172] The energy storage virtual subset establishment module 240 is used to divide the energy storage battery cluster into multiple virtual subsets in the logic control layer of the energy management system when the natural pause time of the time domain matching result characterization is less than the effective recovery window, and to maintain that at least one virtual subset is in a discharge state at any given time.

[0173] The optimal control sequence solving module 250 is used to construct a multi-objective cooperative scheduling model with the objective functions of minimizing the cumulative active voltage and maximizing the effective discharge capacity, and to use an optimization algorithm to solve for the optimal control sequence of the multi-objective cooperative scheduling model.

[0174] The energy storage pulse coordinated control module 260 is used to perform power allocation and state switching operations of each virtual subset through the power converter according to the optimal control sequence to complete the energy storage pulse coordinated control.

[0175] The specific functions and examples of each module and submodule of the device in this embodiment of the invention can be found in the relevant descriptions of the corresponding steps in the above method embodiments, and will not be repeated here.

[0176] According to embodiments of the present invention, the above-described method of the present invention can be applied to an electronic device and a readable storage medium.

[0177] Figure 7 A schematic block diagram of an example electronic device 600 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0178] like Figure 7 As shown, the electronic device 600 includes a computing unit 601, which can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. The RAM 603 may also store various programs and data required for the operation of the electronic device 600. The computing unit 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0179] Multiple components in electronic device 600 are connected to I / O interface 605, including: input unit 606, such as keyboard, mouse, etc.; output unit 607, such as various types of displays, speakers, etc.; storage unit 608, such as disk, optical disk, etc.; and communication unit 609, such as network card, modem, wireless transceiver, etc. Communication unit 609 allows electronic device 600 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0180] The computing unit 601 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs the various methods and processes described above, such as a method for coordinated control of energy storage pulses based on electrochemical recovery. For example, in some embodiments, a method for coordinated control of energy storage pulses based on electrochemical recovery can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 600 via ROM 602 and / or communication unit 609. When the computer program is loaded into RAM 603 and executed by the computing unit 601, one or more steps of the method for coordinated control of energy storage pulses based on electrochemical recovery described above can be performed. Alternatively, in other embodiments, the computing unit 601 may be configured, by any other suitable means (e.g., by means of firmware), to perform a storage pulse cooperative control method based on electrochemical recovery.

[0181] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0182] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0183] In the context of this invention, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0184] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0185] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0186] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0187] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

[0188] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for coordinated control of energy storage pulses based on electrochemical recovery, characterized in that, include: The pulse characteristic parameters of the load are obtained through a high-frequency data acquisition device; Based on the pulse characteristic parameters, an electrochemical relaxation time constant model is constructed using the principles of electrochemical kinetics. The battery operating parameters are input into the electrochemical relaxation time constant model. The effective recovery window of the battery is calculated through the electrochemical relaxation time constant model. The natural pause time in the pulse characteristic parameters is compared with the effective recovery window to generate a time-domain matching result. In response to the time-domain matching result indicating that the natural pause time is less than the effective recovery window, the energy storage battery cluster is divided into multiple virtual subsets in the logic control layer of the energy management system, and at least one of the virtual subsets is in a discharge state at any given time. A multi-objective cooperative scheduling model is constructed with the objectives of minimizing the cumulative active voltage and maximizing the effective discharge capacity as objective functions, and the optimal control sequence of the multi-objective cooperative scheduling model is obtained by using an optimization algorithm. According to the optimal control sequence, the power allocation and state switching operations of each virtual subset are executed by the power converter to complete the energy storage pulse coordinated control.

2. The method according to claim 1, characterized in that, The high-frequency data acquisition device is equipped with a current monitoring unit, a power sampling unit, and a feature extraction unit; the acquisition of the load's pulse characteristic parameters through the high-frequency data acquisition device includes: The current monitoring unit acquires real-time monitoring data of load current ripple. The power sampling unit uses a sliding time window algorithm to discretize the load power to obtain load power sampling data. The feature extraction unit identifies abrupt changes in load power based on the load power sampling data. These abrupt changes include the pulse start time and the pulse end time. The duration of a single pulse is calculated based on the pulse start time and the pulse end time. Calculate the natural pause time based on the pulse start time and pulse end time of adjacent pulses; Based on the load power sampling data, the pulse peak power is extracted; The pulse characteristic parameters are generated based on the single pulse duration, the natural pause time, and the pulse peak power.

3. The method according to claim 2, characterized in that, The electrochemical relaxation time constant model constructed based on the pulse characteristic parameters and combined with the principles of electrochemical kinetics includes: Based on the pulse peak power and single pulse duration in the pulse characteristic parameters, and combined with the battery operating parameters, a second-order resistive-capacitive equivalent circuit model of the battery is constructed. The second-order resistive-capacitive equivalent circuit model includes ohmic internal resistance, a first-order resistive-capacitive circuit characterizing electrochemical polarization, and a second-order resistive-capacitive circuit characterizing concentration polarization. Based on Fick's diffusion law, the diffusion process of lithium ions in the solid phase particles of the electrode is analyzed. Based on the diffusion process and the battery operating parameters, the electrochemical relaxation time constant of the battery is derived to construct the electrochemical relaxation time constant model.

4. The method according to claim 3, characterized in that, The step of calculating the effective recovery window of the battery using the electrochemical relaxation time constant model and comparing the natural pause time in the pulse characteristic parameters with the effective recovery window to generate a time-domain matching result includes: Based on the electrochemical relaxation time constant, the effective recovery window value of the battery is calculated using the recovery window calculation algorithm configured by the electrochemical relaxation time constant model. The temporal matching result is generated by comparing the natural pause time with the effective recovery window. The recovery window calculation algorithm is configured to calculate the shortest resting time required for the battery terminal voltage to recover to a specific proportion of the steady-state value defined by the electrochemical relaxation time constant model, based on the electrochemical relaxation time constant.

5. The method according to claim 2, characterized in that, The step of dividing the energy storage battery cluster into multiple virtual subsets in the logic control layer of the energy management system and maintaining that at least one of the virtual subsets is in a discharged state at any given time includes: Calculate the minimum number of groups based on the natural pause time, the effective recovery window, and the duration of a single pulse. Based on the minimum number of groups, the energy storage battery cluster is divided into multiple virtual subsets; The phase difference between each virtual subset is set so that each virtual subset discharges alternately, thereby maintaining that at least one virtual subset is in a discharging state at any given time.

6. The method according to claim 1, characterized in that, The construction of a multi-objective cooperative scheduling model with the objective functions of minimizing the cumulative active voltage and maximizing the effective discharge capacity, and the use of optimization algorithms to solve for the optimal control sequence of the multi-objective cooperative scheduling model, includes: Define the objective function of the multi-objective cooperative scheduling model, which includes minimizing the cumulative optimized voltage and maximizing the effective discharge capacity; Safety constraints are set for the multi-objective cooperative scheduling model, including battery terminal voltage constraints, temperature rise rate constraints, and power balance constraints. The optimization algorithm is used to numerically solve the multi-objective cooperative scheduling model with the objective function and the security constraints to generate the optimal control sequence; The optimal control sequence includes the switching state sequence and power allocation ratio sequence of each virtual subset.

7. The method according to claim 1, characterized in that, The step of performing power allocation and state switching operations on each of the virtual subsets through a power converter according to the optimal control sequence to complete the energy storage pulse coordinated control includes: The optimal control sequence is sent to the power converter via a communication bus; The power converter controls each virtual subset to switch between the pressurized discharge state and the forced relaxation state according to the switching state sequence and the power allocation ratio sequence in the optimal control sequence. Monitor the terminal voltage recovery curves of each virtual subset; In response to the detection of a sudden change in load characteristic parameters, the process of re-extracting the pulse characteristic parameters and recombining the virtual subset is triggered.

8. A storage pulse coordinated control system based on electrochemical recovery, characterized in that, include: The pulse characteristic parameter acquisition module is used to acquire the pulse characteristic parameters of the load through a high-frequency data acquisition device. The time constant model construction module is used to construct an electrochemical relaxation time constant model based on the pulse characteristic parameters and the principles of electrochemical kinetics. The time-domain matching result generation module is used to input the battery operating parameters into the electrochemical relaxation time constant model, calculate the effective recovery window of the battery through the electrochemical relaxation time constant model, and compare the natural pause time in the pulse characteristic parameters with the effective recovery window to generate a time-domain matching result. The energy storage virtual subset establishment module is used to divide the energy storage battery cluster into multiple virtual subsets in the logic control layer of the energy management system when the time domain matching result indicates that the natural pause time is less than the effective recovery window, and to maintain that at least one of the virtual subsets is in a discharge state at any given time. The optimal control sequence solving module is used to construct a multi-objective cooperative scheduling model with the objective functions of minimizing the cumulative active voltage and maximizing the effective discharge capacity, and to use an optimization algorithm to solve for the optimal control sequence of the multi-objective cooperative scheduling model. The energy storage pulse coordinated control module is used to perform power allocation and state switching operations on each of the virtual subsets through the power converter according to the optimal control sequence to complete the energy storage pulse coordinated control.

9. An electronic device, characterized in that, include: At least one processor; and a memory that is communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, in, Computer instructions are used to cause a computer to perform the method according to any one of claims 1-7.

Images