SOC estimation method and system for balanced charge and discharge control of large-scale energy storage power station

Abstract

The invention provides an SOC estimation method and system for balanced charging and discharging control of a large-scale energy storage power station, and relates to the technical field of battery management systems.The SOC estimation method comprises the steps that working parameters of a single battery are collected, a state model and an observation model are established through a Kalman filtering algorithm, and the SOC of the single battery in the next time step is predicted; historical working parameters and SOC of a sample battery are collected, and a sample data set is generated; fusing the sample data set to correct the predicted SOC, setting a single battery charge state threshold, generating an effective single battery set, and calculating the predicted SOC of the battery module; setting a module charge state threshold threshold, screening effective battery modules, and calculating a battery cluster prediction SOC (State of Charge); setting a battery cluster charge state threshold threshold, screening effective battery clusters, and generating a power station available charge state; according to the effective single battery set, the effective module set and the effective battery cluster set, on-off of a battery discharging loop is controlled, the available charge state of the power station is input into a battery management system, and a discharging strategy is adjusted in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of battery management systems, and specifically to a SOC estimation method and system for balanced charge and discharge control in large-scale energy storage power stations. Background Art

[0002] With the widespread adoption of renewable energy and the growing demand for grid stability, large-scale energy storage power stations, as a crucial component of the energy system, are crucial for the safe and stable operation of the power system. In the operation and management of energy storage power stations, accurate estimation of the battery state of charge (SOC) is a key technology for achieving balanced charge and discharge control, directly impacting the service life and operational efficiency of the energy storage system.

[0003] At present, the estimation methods of battery SOC in energy storage power stations mainly include open circuit voltage method, ampere-hour integration method, Kalman filter method, etc. Among them, the Kalman filter method has been widely used in the field of battery SOC estimation because it can effectively handle system noise and measurement errors. For example, the prior art of publication number CN114239463A discloses a battery cluster state of charge correction method based on big data. This method improves the SOC estimation accuracy by establishing a battery cluster equivalent circuit model, using an adaptive harmonic search algorithm to identify model parameters, and combining the unscented Kalman filter algorithm to correct the battery cluster SOC. In terms of multi-level SOC aggregation, CN113608130A proposes an online estimation method for battery cluster state of charge. This method first calculates the state of charge of the single cell, then calculates the state of charge of the battery stack, and finally calculates the battery cluster state of charge by combining the single cell state of charge and the battery stack state of charge, thereby realizing real-time hierarchical calculation of the state of charge of the battery stack, battery cluster and single cell. This hierarchical estimation method helps improve the accuracy of battery cluster SOC calculations, enabling the operating status of battery stacks and clusters to track the charge and discharge status of individual cells. However, existing SOC estimation methods in battery management systems still suffer from the following issues: Individual cell SOC estimation accuracy is insufficient. Existing methods often rely on a single estimation algorithm, such as the Kalman filter or the ampere-hour integration method, which struggles to adapt to the dynamic characteristics of batteries under varying operating conditions, leading to deviations between the estimated results and the actual state of charge. Especially in large-scale energy storage power plants, where the number of individual cells is large, the consistency differences between individual cells are more pronounced, further exacerbating the difficulty of SOC estimation. Secondly, there is a lack of effective multi-level SOC aggregation mechanisms. While existing hierarchical estimation methods exist, most only consider simple arithmetic or weighted averages, failing to fully account for inter-cell variability, resulting in inaccurate overall SOC calculations for battery modules. In large-scale energy storage power plants, the multi-level SOC aggregation process, from individual cells to modules, battery clusters, and ultimately the entire power plant, remains a challenge in effectively handling information transfer and state fusion across different levels. Thirdly, existing methods fail to fully account for inter-cell variability and dynamic changes. Due to factors such as manufacturing process, operating environment, and degree of aging, battery cells within the same energy storage power station often exhibit performance differences, and these differences change dynamically with charge and discharge cycles. Existing methods lack effective mechanisms for identifying and addressing these differences, which can easily lead to overcharging or over-discharging of some battery cells, affecting system safety and service life. Finally, there is a lack of an SOC correction mechanism based on historical data. Existing SOC estimation methods are mostly based on real-time measurement data and theoretical models, failing to fully utilize the battery characteristic information contained in historical operating data. They are unable to adapt to the SOC estimation needs under complex operating conditions such as battery aging, affecting the accuracy and reliability of the estimation results. Summary of the Invention

[0004] The purpose of the present invention is to provide a SOC estimation method and system for balanced charge and discharge control of a large-scale energy storage power station, so as to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions: A SOC estimation method for balanced charge and discharge control of a large-scale energy storage power station, comprising the following steps: S1: Collect the real-time operating parameters of each single battery in the energy storage power station to be estimated, and establish a state model and observation model based on the Kalman filter algorithm to predict the predicted SOC value of each single battery in the next time step; S2: Obtain the historical operating parameters of each single battery in the energy storage power station and the SOC data corresponding to the next time step, and map them to form a set of reference vectors. All reference vectors are formed into a mapping relationship library. A similarity criterion is defined. The mapping relationship library is matched based on the real-time operating parameters of each single battery collected. Reference vectors that meet the similarity criterion are selected, and the SOC data in each reference vector is extracted to form a sample data set. S3: Use the SOC data in the sample data set corresponding to each single cell to correct the predicted SOC for the next time step, obtain the corrected SOC value of each single cell for the next time step, and calibrate it as the reference value. Set the threshold value of the single cell state of charge and screen out the valid single cell set based on the relationship between the reference value and the threshold value. Calculate the SOC prediction value of the battery module to which it belongs according to the rated capacity weight. S4: Based on the predicted SOC of the battery module, the module state of charge threshold is set, and a valid battery module set is screened within the battery cluster. The SOC prediction value of the battery cluster to which it belongs is calculated weighted by the module capacity. Based on the predicted SOC of the battery cluster, the battery cluster state of charge threshold is set, and a valid battery cluster set is screened at the energy storage power station level. The SOC prediction value of the power station is calculated weighted by the cluster capacity. S5: Execute three-level discharge loop control based on the effective single battery set, effective module set and effective battery cluster set; input the power station SOC prediction value into the battery management system, and adjust the discharge strategy in real time to ensure that low-power cells stop discharging and high-power cells continue to supply power.

[0006] Furthermore, the real-time operating parameters of each single cell in the energy storage power station to be estimated are collected, including initial state of charge, current, voltage, ohmic internal resistance, and polarization resistance. A state model is established based on the Kalman filter algorithm as follows: ; in, for Moment single battery The predicted SOC value, for Moment single battery The predicted value of polarization voltage, is the rated capacity of the single battery to be estimated, is the time step, is the polarization time constant, is the Coulomb efficiency, is the polarization resistance of the single cell to be estimated, For single battery Current at each moment, is the process noise; establish the observation equation and determine the physical relationship between the state equation and the observation data. The formula is as follows: ; in, For batteries The actual measured voltage between the positive and negative electrodes, is the slope of the single cell open circuit voltage to SOC, is the ohmic internal resistance, is the observation noise.

[0007] Furthermore, the historical working parameters of the single battery in the energy storage power station at each moment and the SOC data corresponding to the next time step are obtained, and mapped to form a reference vector. All reference vectors are constructed into a mapping relationship library, where the historical working parameters are , is the discharge current value at time k, is the terminal voltage at the positive and negative terminals at time k, is the operating temperature of the battery at time k, is the state of charge at time k; define the similarity criterion, match the real-time working parameters of each single battery collected in the mapping relationship library, filter out the reference vectors that meet the similarity criterion, extract the SOC data in each reference vector, and form a sample data set; the similarity criterion requires sample batteries The following conditions are met between the discharge operating parameters of the single battery to be tested: ; in, For sample batteries Discharge current, Single battery to be tested Discharge current, For sample batteries Rated capacitance, Sample battery Current operating temperature, Single battery to be tested Operating temperature, Sample battery Current SOC, Single battery to be tested Current SOC; use similarity criteria to filter out sample data sets in the mapping relationship library ,in is the battery reference vector that satisfies the similarity criterion.

[0008] Furthermore, the SOC data in the sample data set corresponding to each single battery is used to correct the predicted SOC of the next time step to generate the corrected battery State vector, obtain the corrected SOC value of each single battery in the next time step and calibrate it as the reference value: ; in, for The SOC reference value of the single battery after constant correction, for Moment single battery The corrected predicted polarization voltage, For single battery after time correction The Kalman gain matrix, is the observation matrix, For batteries in the dataset that meet the similarity criteria The gain matrix between Batteries in the dataset that satisfy the similarity criterion The voltage observation value, For sample batteries The observation matrix, for Sample battery SOC value, for Time sample battery polarization voltage value.

[0009] Furthermore, a threshold value of the state of charge of a single battery is set and a valid single battery set is screened out according to the relationship between the reference value and the threshold value, and the SOC prediction value of the battery module to which it belongs is calculated weighted by the rated capacity. ; in, For the effective single battery collection, The SOC threshold of the single battery is calculated by weighting the rated capacity to calculate the SOC prediction value of the battery module to which it belongs: ; in, for Moment battery module SOC, For single battery Rated capacitance, get module Predict SOC.

[0010] Furthermore, based on the predicted SOC of the battery module, the module state of charge threshold is set, and the valid battery module set is screened in the battery cluster. The SOC prediction value of the battery cluster to which it belongs is calculated weighted by the module capacity: ; in, For the effective battery module collection, The battery module state of charge threshold is used to calculate the battery cluster SOC: ; in, for Time battery cluster SOC, For effective single cell Module The rated capacitance is based on the predicted SOC of the battery cluster. The battery cluster state of charge threshold is set. Valid battery clusters are screened at the energy storage power station level, and the predicted SOC value of the power station is calculated weighted by cluster capacity: ; in, is the effective battery cluster set, for The SOC of the energy storage power station at all times, is the battery cluster state of charge threshold, For valid modules Battery cluster Effective rated capacitance.

[0011] Furthermore, according to the effective single battery set, effective module set and effective battery cluster set, a three-level discharge circuit control is executed: Single discharge circuit, isolated Module contactor, close The cluster PCS converter inputs the power station's available state of charge, as well as the charge state of each battery cluster, battery module, and single battery into the battery management system, and adjusts the discharge strategy in real time to ensure that low-charge cells stop discharging while high-charge cells continue to supply power.

[0012] The present invention also provides an SOC estimation system for balanced charge and discharge control of a large-scale energy storage power station. The system is used to implement the above-mentioned SOC estimation method for balanced charge and discharge control of a large-scale energy storage power station, specifically comprising: Single cell SOC prediction module: This module collects the real-time operating parameters of each single cell in the energy storage power station to be estimated, and uses the Kalman filter algorithm to establish a state model and observation model to predict the predicted SOC value of each single cell in the next time step. Sample data acquisition module: This module is used to obtain the historical operating parameters of each single battery in the energy storage power station and the SOC data corresponding to the next time step, and then map them to form a set of reference vectors. All reference vectors are combined into a mapping relationship library, and similarity criteria are defined. The mapping relationship library is matched based on the real-time operating parameters of each single battery collected. Reference vectors that meet the similarity criteria are selected, and the SOC data in each reference vector is extracted to form a sample data set. SOC correction and estimation module: used to use the SOC data in the sample data set corresponding to each single cell to correct the predicted SOC for the next time step, obtain the corrected SOC value of each single cell in the next time step, and calibrate it as the reference value, set the threshold value of the single cell state of charge, and screen out the valid single cell set based on the relationship between the reference value and the threshold value, and calculate the SOC prediction value of the battery module to which it belongs according to the rated capacity weight; Energy storage power station SOC estimation module: This module is used to predict the SOC based on the battery module, set the module state of charge threshold, screen the valid battery module set within the battery cluster, and calculate the SOC prediction value of the battery cluster weighted by the module capacity. Based on the battery cluster SOC prediction, the battery cluster state of charge threshold is set, and the valid battery cluster set is screened at the energy storage power station level, and the power station SOC prediction value is weighted by the cluster capacity. Discharge control and strategy adjustment module: Executes three-level discharge loop control based on the effective single battery set, effective module set, and effective battery cluster set; inputs the power station SOC prediction value into the battery management system and adjusts the discharge strategy in real time to ensure that low-power cells stop discharging and high-power cells continue to supply power.

[0013] Compared with the prior art, the present invention has the following beneficial effects: Through the Kalman filter and historical sample data fusion correction mechanism, the accuracy of single-cell battery SOC prediction is significantly improved, and the accuracy is higher than that of traditional methods; a four-level SOC calculation system of single-cell-module-cluster-power station is established to ensure that the SOC calculation at each level accurately reflects the available energy, solving the problem of lack of multi-level aggregation in traditional methods; through the 10% SOC threshold value throughout the single-cell, module, and cluster step-by-step shutdown mechanism, the battery over-discharge damage is effectively avoided and the battery service life is extended; the strategy of prioritizing the use of high-capacity cells extends the overall battery service life; the three-level shutdown mechanism ensures that the system is safe and controllable under abnormal conditions, prevents safety accidents, and improves the overall safety of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 Schematic diagram of the overall method flow of the present invention; Figure 2 This is a comparison chart of the SOC estimation experimental data of the present invention; Figure 3 It is a schematic block diagram of the SOC estimation system in the present invention. DETAILED DESCRIPTION

[0015] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0016] It should be noted that, unless otherwise defined, the technical or scientific terms used in the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0017] Example: See also Figure 1-2 , the present invention provides a technical solution: A SOC estimation method for balanced charge and discharge control of a large-scale energy storage power station, comprising the following steps: S1: Collect the real-time operating parameters of each single battery in the energy storage power station to be estimated, and establish a state model and observation model based on the Kalman filter algorithm to predict the predicted SOC value of each single battery in the next time step; In this embodiment, the real-time operating parameters of each single cell in the energy storage power station to be estimated are collected, including: initial state of charge, current, voltage, ohmic internal resistance, and polarization resistance. A state model is established based on the Kalman filter algorithm as follows: ; in, for Moment single battery The predicted SOC value, for Moment single battery The predicted value of polarization voltage, is the rated capacity of the single battery to be estimated, is the time step, is the polarization time constant, is the Coulomb efficiency, is the polarization resistance of the single cell to be estimated, For single battery Current at each moment, is the process noise; establish the observation equation and determine the physical relationship between the state equation and the observation data. The formula is as follows: ; in, For batteries The actual measured voltage between the positive and negative electrodes, is the slope of the single cell open circuit voltage to SOC, is the ohmic internal resistance, The traditional ampere-hour integration method is easily affected by the cumulative error of current noise. This scheme significantly reduces noise interference through bidirectional correction of the state space model and observation equation. Process noise is a random vector that represents the uncertainty of the state prediction model itself and the unmodeled dynamic characteristics. It covers all interference factors that affect the change of state (SOC and polarization voltage) but cannot be accurately described by deterministic equations. Most schemes only use the single state variable SOC, ignore the hysteresis effect of polarization voltage, and introduce the state equation into the process noise. , = + , explicitly describes the double-layer capacitor charging and discharging process, accurately depicts the voltage relaxation effect during current step, and reduces the SOC estimation error under discharge conditions; the observation equation is embedded , the SOC difference is amplified by the slope term to solve the problem of insufficient voltage resolution in the low / high SOC area; the next time step in the predicted SOC value of each single cell at the next time step refers to the current moment After a discrete time points after , rather than absolute physical time; the time step unifies the calculation cycle of all monomers to avoid timing confusion caused by acquisition delay; In this embodiment, two sets of discharge simulation experiments were conducted on the same energy storage power station to compare this technical solution with the traditional SOC estimation solution. The data simulates a 3MWh energy storage power station (containing 40 battery clusters, 20 modules per cluster, and 24 NMC lithium batteries per module). The experimental data of the single cell in battery module B12 at the 115-minute time is extracted as shown in the following table: Table 1: Kalman filter prediction of single-unit SOC data table

[0018] S2: Obtain the historical operating parameters of each single battery in the energy storage power station and the SOC data corresponding to the next time step, and map them to form a set of reference vectors. All reference vectors are formed into a mapping relationship library. A similarity criterion is defined. The mapping relationship library is matched based on the real-time operating parameters of each single battery collected. Reference vectors that meet the similarity criterion are selected, and the SOC data in each reference vector is extracted to form a sample data set. In this embodiment, the historical operating parameters of the single battery in the energy storage power station at each moment and the SOC data corresponding to the next time step are obtained, and mapped to form a set of reference vectors. All reference vectors are used to form a mapping relationship library, where the historical operating parameters are , is the discharge current value at time k, is the terminal voltage at the positive and negative terminals at time k, is the operating temperature of the battery at time k, is the state of charge at time k; define a similarity criterion, match the collected real-time operating parameters of each single battery in the mapping relationship library, filter out reference vectors that meet the similarity criterion, extract the SOC data in each reference vector, and form a sample data set; the similarity criterion requires that the discharge operating parameters of the sample battery and the single battery to be tested meet the following conditions: ; in, For sample batteries Discharge current, Single battery to be tested Discharge current, For sample batteries Rated capacitance, Sample battery Current operating temperature, Single battery to be tested Operating temperature, Sample battery Current SOC, Single battery to be tested Current SOC; use similarity criteria to filter out sample data sets in the mapping relationship library ,in is the battery reference vector that satisfies the similarity criterion; Using the physical basis of three-dimensional similarity to define similarity criteria, current constraints require historical sample batteries With target battery The discharge current difference is less than 10% of the rated capacity to avoid the polarization effect caused by the difference in large current working conditions; the temperature constraint requires that the historical sample battery With target battery The operating temperature is within ±2°C, covering the nonlinear range of lithium ion diffusion rate and reaction activity; SOC constraint requires historical sample batteries With target battery The SOC of the sample battery is less than 5% of the window, ensuring that the operating point is in the similar slope section of the OCV-SOC curve; compared with the direct training of the neural network without similarity screening, the prediction error at the boundary condition is reduced, and the sample battery With target battery The similarity criterion is the core condition for screening historical sample data and is used to correct the initial SOC prediction value of the Kalman filter algorithm; Taking the target battery B1203 as an example, its parameters at 115 minutes are , , , the sample data set is filtered out as follows: Table 2: Battery B1203 sample data set data table

[0019] S3: Use the SOC data in the sample data set corresponding to each single cell to correct the predicted SOC for the next time step, obtain the corrected SOC value of each single cell for the next time step, and calibrate it as the reference value. Set the threshold value of the single cell state of charge and screen out the valid single cell set based on the relationship between the reference value and the threshold value. Calculate the SOC prediction value of the battery module to which it belongs according to the rated capacity weight. In this embodiment, the SOC data in the sample data set corresponding to each single battery is used to correct the predicted SOC of the next time step to generate the corrected battery State vector, obtain the corrected SOC value of each single battery in the next time step and calibrate it as the reference value: ; in, for The SOC reference value of the single battery after constant correction, for Moment single battery The corrected predicted polarization voltage, For single battery after time correction The Kalman gain matrix, is the observation matrix, , For batteries in the dataset that meet the similarity criteria The gain matrix between Batteries in the dataset that satisfy the similarity criterion The voltage observation value, For sample batteries The observation matrix, for Sample battery SOC value, for Sample battery polarization voltage value.

[0020] Correct current battery Self-observation residuals to solve the model uncertainty of the current battery; ,in Battery With sample The historical covariance of SOC, sample SOC historical variance; reflects the battery With sample battery The mutual influence relationship between the states is essentially an amplification factor. The observed residuals Acts proportionally on the target battery The correction term uses the sample data set to compensate for the Kalman filter algorithm deviation; in this embodiment, the corrected data is as follows: Table 3: Corrected single-cell SOC data table

[0021] By integrating real-time observations with historical data on similar operating conditions, the model mismatch problem of Kalman filtering during battery aging and temperature mutations can be overcome. Actual sample data can be used to correct the initial estimate and avoid misjudgment of low-battery batteries. For example, when the SOC approaches 10%, invalid cells can be accurately disconnected. Sample data is continuously updated to automatically track battery performance degradation without the need for repeated calibration of model parameters.

[0022] In this embodiment, a threshold value of the state of charge of a single battery is set to generate a valid single battery set. ; in, For the effective single battery collection, is the state of charge threshold of the single cell, and the weighted average of the valid single cells in each module is calculated: ; in, for Moment battery module SOC, For single battery Rated capacitance, get module Predict SOC.

[0023] The design concept of this solution represents that during the discharge process of the energy storage power station, invalid single cells are excluded and not included in the estimated SOC value. If the estimated value is included, the total number of single cells in the battery module will be greater than the actual number of dischargeable single cells, which may easily cause over-discharge, shorten the battery life, and affect the efficiency of the energy storage power station. At the same time, it also reflects the principle of large-capacity battery dominance, preventing abnormal values ​​of small-capacity batteries from affecting the overall SOC, and at the same time avoiding deep discharge of single cells.

[0024] S4: Based on the battery module SOC prediction, set the module SOC threshold, screen the valid battery module set within the battery cluster, and calculate the SOC prediction value of the battery cluster to which it belongs by weighted module capacity; based on the battery cluster SOC prediction, set the battery cluster SOC threshold, screen the valid battery cluster set at the energy storage power station level, and calculate the SOC prediction value of the power station by weighted cluster capacity: In this embodiment, based on the battery module predicted SOC, the module state of charge threshold is set, and a valid battery module set is screened within the battery cluster. The SOC predicted value of the battery cluster to which it belongs is calculated weighted by the module capacity: ; in, For the effective battery module collection, The battery module state of charge threshold is used to calculate the battery cluster SOC: ; in, for Time battery cluster SOC, For effective single cell Module Rated capacitance; This formula strictly follows the law of conservation of charge, the total available charge of the battery cluster Total capacity It truly reflects energy reserves rather than numerical averages. Compared with traditional methods, when there are batteries with capacity decay in the battery cluster, their SOC is inflated due to increased internal resistance. The arithmetic average of the traditional scheme will cause serious distortion of SOC estimation and higher error. This scheme calculates the SOC of the battery cluster by considering only the effective battery modules and only sums the dischargeable battery modules.

[0025] In this embodiment, based on the predicted SOC of the battery cluster, a battery cluster state of charge threshold is set, a valid battery cluster set is screened at the energy storage power station level, and the power station SOC prediction value is calculated weighted by cluster capacity: ; in, is the effective battery cluster set, for The SOC of the energy storage power station at all times, is the battery cluster state of charge threshold, For valid modules Battery cluster Effective rated capacitance.

[0026] Similar to the screening of single-point batteries and battery modules, it realizes the "step-by-step attenuation" management of available capacity from single cells to power stations, and abandons the ineffective battery level; In this embodiment, two discharge simulation experiments were conducted on the same energy storage power station, that is, a comparative experiment was conducted between this technical solution and a traditional SOC estimation solution; the following experimental data were obtained: as shown in the following table: Table 4: Experimental data table

[0027] Experimental data shows that at the end of the test, there were 6,610 valid cells, accounting for 34.4% of the total capacity, and they were still working. The average SOC of the 12,590 isolated cells was only 7.2%. Through precise isolation, this technical solution enabled healthy cells to release 21.5% more energy than the discharge instructions ultimately affected by the traditional SOC estimation solution. The predicted value of the traditional solution caused the error to deteriorate exponentially in the middle and late stages of discharge, causing the system to terminate discharge prematurely. This technical solution, through Kalman filtering and sample fusion, controlled the error within ±3%. Due to the active isolation of low-power cells, the remaining high-SOC cells continued to discharge. Although a negative error was shown (-12.29% at 150 minutes), this was additional capacity released by system reconstruction, and the available SOC range was expanded from 23% of the traditional solution to 31.5%, releasing 44.1% more energy.

[0028] See Figure 2 , Figure 2 The deviation and absolute error between the SOC estimates of the traditional solution and this technical solution and the actual value of the energy storage power station in Table 4 are clearly shown. It can be seen that the deviation of this technical solution from the actual value is smaller than that of the traditional technical solution. In addition, during the experiment, this technical solution screened the effective battery levels. Due to the accurate estimation, this technical solution makes the discharge capacity of the energy storage power station larger and safer. The 10% SOC threshold is used throughout the step-by-step shutdown mechanism of the single cell, module, and cluster, which effectively avoids battery over-discharge damage and extends the battery life. The strategy of prioritizing the use of high-capacity cells prolongs the overall battery life. The three-level shutdown mechanism ensures that the system is safe and controllable under abnormal conditions.

[0029] S5: Execute three-level discharge loop control based on the effective single battery set, effective module set and effective battery cluster set; input the power station SOC prediction value into the battery management system, and adjust the discharge strategy in real time to ensure that low-power cells stop discharging and high-power cells continue to supply power.

[0030] In this embodiment, a three-level discharge circuit control is performed based on the effective single battery set, the effective module set and the effective battery cluster set: Single discharge circuit, isolated Module contactor, close The cluster PCS converter is integrated with the power plant's available state of charge (SOC), as well as the SOC of each battery cluster, battery module, and single cell, into the battery management system. This allows for real-time adjustment of discharge strategies, ensuring that low-charge cells cease discharging while high-charge cells continue to supply power. The power plant's available SOC is input into the energy management system, enabling adaptive adjustment of discharge power. Intelligent scheduling prioritizes discharge of high-charge clusters, extending system life. A three-level hardware shutdown and a 10% SOC hard threshold completely eliminate the risk of thermal runaway caused by over-discharge, improving available capacity utilization. See also Figure 3 The present invention also provides an SOC estimation system for balanced charge and discharge control of a large-scale energy storage power station. The system is used to implement the above-mentioned SOC estimation method for balanced charge and discharge control of a large-scale energy storage power station, specifically comprising: Single cell SOC prediction module: This module collects the real-time operating parameters of each single cell in the energy storage power station to be estimated, and uses the Kalman filter algorithm to establish a state model and observation model to predict the predicted SOC value of each single cell in the next time step. Sample data acquisition module: This module is used to obtain the historical operating parameters of each single battery in the energy storage power station and the SOC data corresponding to the next time step, and then map them to form a set of reference vectors. All reference vectors are combined into a mapping relationship library, and similarity criteria are defined. The mapping relationship library is matched based on the real-time operating parameters of each single battery collected. Reference vectors that meet the similarity criteria are selected, and the SOC data in each reference vector is extracted to form a sample data set. SOC correction and estimation module: used to use the SOC data in the sample data set corresponding to each single cell to correct the predicted SOC for the next time step, obtain the corrected SOC value of each single cell in the next time step, and calibrate it as the reference value, set the threshold value of the single cell state of charge, and screen out the valid single cell set based on the relationship between the reference value and the threshold value, and calculate the SOC prediction value of the battery module to which it belongs according to the rated capacity weight; Energy storage power station SOC estimation module: This module is used to predict the SOC based on the battery module, set the module state of charge threshold, screen the valid battery module set within the battery cluster, and calculate the SOC prediction value of the battery cluster weighted by the module capacity. Based on the battery cluster SOC prediction, the battery cluster state of charge threshold is set, and the valid battery cluster set is screened at the energy storage power station level, and the power station SOC prediction value is weighted by the cluster capacity. Discharge control and strategy adjustment module: used to perform three-level discharge loop control based on the effective single battery set, effective module set and effective battery cluster set; input the power station SOC prediction value into the battery management system, and adjust the discharge strategy in real time to ensure that low-power cells stop discharging and high-power cells continue to supply power.

[0031] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0032] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraints of the technical solution.

[0033] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment as needed.

[0034] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A SOC estimation method for balanced charge and discharge control of a large-scale energy storage power station, characterized in that: The specific steps include: S1: Collect the real-time operating parameters of each single battery in the energy storage power station to be estimated, and establish a state model and observation model based on the Kalman filter algorithm to predict the predicted SOC value of each single battery in the next time step; S2: Obtain the historical operating parameters of each single battery in the energy storage power station and the SOC data corresponding to the next time step, and map them to form a set of reference vectors. All reference vectors are formed into a mapping relationship library. A similarity criterion is defined. The mapping relationship library is matched based on the real-time operating parameters of each single battery collected. Reference vectors that meet the similarity criterion are selected, and the SOC data in each reference vector is extracted to form a sample data set. S3: Use the SOC data in the sample data set corresponding to each single cell to correct the predicted SOC for the next time step, obtain the corrected SOC value of each single cell for the next time step, and calibrate it as the reference value. Set the threshold value of the single cell state of charge and screen out the valid single cell set based on the relationship between the reference value and the threshold value. Calculate the SOC prediction value of the battery module to which it belongs according to the rated capacity weight. S4: Based on the predicted SOC of the battery module, the module state of charge threshold is set, and a valid battery module set is screened within the battery cluster. The SOC prediction value of the battery cluster to which it belongs is calculated weighted by the module capacity. Based on the predicted SOC of the battery cluster, the battery cluster state of charge threshold is set, and a valid battery cluster set is screened at the energy storage power station level. The SOC prediction value of the power station is calculated weighted by the cluster capacity. S5: Execute three-level discharge loop control based on the effective single battery set, effective module set and effective battery cluster set; input the power station SOC prediction value into the battery management system, and adjust the discharge strategy in real time to ensure that low-power cells stop discharging and high-power cells continue to supply power.

2. The SOC estimation method for balanced charge and discharge control of a large-scale energy storage power station according to claim 1, characterized in that: The real-time operating parameters of each single cell in the energy storage power station to be estimated are collected, including initial state of charge, current, voltage, ohmic internal resistance, and polarization resistance. The state model is established based on the Kalman filter algorithm as follows: ； in, for Moment single battery The predicted SOC value, for Moment single battery The predicted value of polarization voltage, is the rated capacity of the single battery to be estimated, is the time step, is the polarization time constant, is the Coulomb efficiency, is the polarization resistance of the single cell to be estimated, For single battery Current at each moment, is the process noise; establish the observation equation and determine the physical relationship between the state equation and the observation data. The formula is as follows: ； in, For batteries The actual measured voltage between the positive and negative electrodes, is the slope of the single cell open circuit voltage to SOC, is the ohmic internal resistance, is the observation noise.

3. The SOC estimation method for balanced charge and discharge control of a large-scale energy storage power station according to claim 1, characterized in that: Obtain the historical working parameters of each cell in the energy storage power station and the SOC data corresponding to the next time step, and map them to form a set of reference vectors. All reference vectors form a mapping relationship library, where the historical working parameters are , is the discharge current value at time k, is the terminal voltage at the positive and negative terminals at time k, is the operating temperature of the battery at time k, is the state of charge at time k; Define the similarity criterion, match the real-time working parameters of each single battery in the mapping relationship library, filter out the reference vectors that meet the similarity criterion, extract the SOC data in each reference vector, and form a sample data set; the similarity criterion requires sample batteries The following conditions are met between the discharge operating parameters of the single battery to be tested: ； in, For sample batteries Discharge current, Single battery to be tested Discharge current, For sample batteries Rated capacitance, Sample battery Current operating temperature, Single battery to be tested Operating temperature, Sample battery Current SOC, Single battery to be tested Current SOC; use similarity criteria to filter out sample data sets in the mapping relationship library ,in is the battery reference vector that satisfies the similarity criterion.

4. The SOC estimation method for balanced charge and discharge control of a large-scale energy storage power station according to claim 3, characterized in that: Use the SOC data in the sample data set corresponding to each single battery to correct the predicted SOC for the next time step and generate the corrected battery State vector, obtain the corrected SOC value of each single battery in the next time step and calibrate it as the reference value: ； in, for The SOC reference value of the single battery after constant correction, for Moment single battery The corrected predicted polarization voltage, For single battery after time correction The Kalman gain matrix, is the observation matrix, For batteries in the dataset that meet the similarity criteria The gain matrix between Batteries in the dataset that satisfy the similarity criterion The voltage observation value, For sample batteries The observation matrix, for Sample battery SOC value, for Sample battery polarization voltage value.

5. The SOC estimation method for balanced charge and discharge control of a large-scale energy storage power station according to claim 4, characterized in that: Set the threshold value of the charge state of the single cell and screen out the valid single cell set based on the relationship between the reference value and the threshold value, and calculate the SOC prediction value of the battery module to which it belongs according to the rated capacity weight: ； in, For the effective single battery collection, The SOC threshold of the single battery is calculated by weighting the rated capacity to calculate the SOC prediction value of the battery module to which it belongs: ； in, for Moment battery module SOC, For single battery Rated capacitance, get module Predict SOC.

6. The SOC estimation method for balanced charge and discharge control of a large-scale energy storage power station according to claim 5, characterized in that: Based on the battery module predicted SOC, the module state of charge threshold is set, and the valid battery module set is screened in the battery cluster. The SOC prediction value of the battery cluster to which it belongs is calculated by weighting the module capacity: ； in, For the effective battery module collection, The battery module state of charge threshold is used to calculate the battery cluster SOC: ； in, for Time battery cluster SOC, For effective single cell Module Rated capacitance.

7. The SOC estimation method for balanced charge and discharge control of a large-scale energy storage power station according to claim 6, characterized in that: Based on the battery cluster SOC prediction, the battery cluster state of charge threshold is set, and the effective battery cluster set is screened at the energy storage power station level. The power station SOC prediction value is calculated by weighting the cluster capacity: ； in, is the effective battery cluster set, for The SOC of the energy storage power station at all times, is the battery cluster state of charge threshold, For valid modules Battery cluster Effective rated capacitance.

8. The SOC estimation method for balanced charge and discharge control of a large-scale energy storage power station according to claim 7, characterized in that: According to the effective single battery set, effective module set and effective battery cluster set, three-level discharge circuit control is executed: cut off Single discharge circuit, isolated Module contactor, close The cluster PCS converter inputs the power station's available state of charge, as well as the charge state of each battery cluster, battery module, and single battery into the battery management system, and adjusts the discharge strategy in real time to ensure that low-charge cells stop discharging while high-charge cells continue to supply power.

9. A SOC estimation system for balanced charge and discharge control of a large-scale energy storage power station, characterized by: The SOC estimation system for balanced charge and discharge control of a large-scale energy storage power station is used to execute the SOC estimation method according to any one of claims 1 to 8, comprising: Single cell SOC prediction module: This module collects the real-time operating parameters of each single cell in the energy storage power station to be estimated, and uses the Kalman filter algorithm to establish a state model and observation model to predict the predicted SOC value of each single cell in the next time step. Sample data acquisition module: This module is used to obtain the historical operating parameters of each single battery in the energy storage power station and the SOC data corresponding to the next time step, and then map them to form a set of reference vectors. All reference vectors are combined into a mapping relationship library, and similarity criteria are defined. The mapping relationship library is matched based on the real-time operating parameters of each single battery collected. Reference vectors that meet the similarity criteria are selected, and the SOC data in each reference vector is extracted to form a sample data set. SOC correction and estimation module: used to use the SOC data in the sample data set corresponding to each single cell to correct the predicted SOC for the next time step, obtain the corrected SOC value of each single cell in the next time step, and calibrate it as the reference value, set the threshold value of the single cell state of charge, and screen out the valid single cell set based on the relationship between the reference value and the threshold value, and calculate the SOC prediction value of the battery module to which it belongs according to the rated capacity weight; Battery cluster SOC estimation module: used to predict SOC based on battery modules, set module state of charge thresholds, screen valid battery module sets within the battery cluster, and calculate the SOC prediction value of the battery cluster to which it belongs based on the weighted module capacity; Energy storage power station SOC estimation module: used to predict SOC based on battery clusters, set battery cluster state of charge thresholds, screen valid battery cluster sets at the energy storage power station level, and calculate the power station SOC prediction value weighted by cluster capacity; Discharge control and strategy adjustment module: used to perform three-level discharge loop control based on the effective single battery set, effective module set and effective battery cluster set; input the power station SOC prediction value into the battery management system, and adjust the discharge strategy in real time to ensure that low-power cells stop discharging and high-power cells continue to supply power.

Images