Energy storage device and working condition prediction method of energy storage device

Abstract

This application discloses an energy storage device and a method for predicting the operating conditions of the energy storage device. It primarily estimates the energy state of the energy storage device based on a hidden Markov model and an equivalent circuit model of a single battery cell. It fully utilizes historical state information data from the individual battery cells in the energy storage device, combined with an unsupervised learning algorithm, to establish an operating condition prediction model for the energy storage device, achieving accurate prediction of future operating conditions. Furthermore, the predicted future operating condition information sequence, combined with the equivalent circuit model, enables precise estimation of the energy state of the energy storage device.

Description

Technical Field

[0001] This application relates to the field of energy technology, and in particular to an energy storage device and a method for predicting the operating conditions of the energy storage device. Background Technology

[0002] The electrification of transportation is a requirement for energy technology development and the realization of smart, environmentally friendly, and energy-saving travel modes worldwide. Energy storage systems, as a crucial component, have long constrained the development of electric vehicles, electric ships, and electric aircraft. Lithium-ion batteries, with their high energy density, high power output, and long lifespan, are considered the optimal choice for energy storage systems. In the future, lithium-ion batteries are trending towards larger sizes and more modular designs, leading to increasingly stringent requirements for energy state monitoring. The energy state of lithium-ion batteries is crucial for estimating the driving range and energy distribution of energy storage systems. Therefore, it is necessary to establish appropriate energy state prediction methods for lithium-ion batteries to achieve accurate predictions and provide a basis for the operation of battery management systems (BMS).

[0003] Current methods for predicting the state of energy (SEE) of batteries struggle to accurately predict the SEE under various operating conditions, thus limiting their applicability. Therefore, there is an urgent need to find a method that can accurately predict the SEE of batteries under diverse operating conditions. Summary of the Invention

[0004] This application provides an energy storage device and a method for predicting the operating conditions of the energy storage device, so as to achieve a more accurate prediction of the energy state of the energy storage device.

[0005] Firstly, an energy storage device is provided, comprising battery cells and further including a battery management system and a condition prediction and compensation device. The battery management system may include a battery acquisition interface and a processor, while the condition prediction and compensation device may include a data communication module and a historical data storage device. The battery acquisition interface is used to acquire the state information of the battery cells during operation. The processor can be used to store the state information in the historical data storage device. The data communication module can be used to package the state information of the battery cells stored in the historical data storage device during the discharge process from the start to the end of discharge into a training dataset and upload it to a cloud server; and to receive a future condition prediction model trained by the cloud server based on the training dataset. Based on this, the processor can also be used to calculate the future condition information sequence of the energy storage device according to the future condition prediction model and the state information of the battery cells. Using the energy storage device provided in this application, a more accurate prediction of the future condition information sequence of the energy storage device can be achieved through cloud collaboration, thereby effectively improving the energy state of the energy storage device calculated based on the future condition information sequence.

[0006] In one possible implementation of this application, the processor can also be used to calculate the prior energy state of the energy storage device based on the future operating condition information sequence and the state information of the battery cells.

[0007] To improve the prediction accuracy of the energy state of an energy storage device, in one possible implementation of this application, the operating condition prediction compensation device may further include a compensation device. This compensation device can determine the energy state compensation value of the energy storage device based on a pre-established temperature compensation model and the predicted future operating condition information sequence of the energy storage device. Then, the processor can calculate the energy state of the energy storage device based on the aforementioned prior energy state and the energy state compensation value.

[0008] Furthermore, during a discharge cycle from start to finish, the state information of individual battery cells changes significantly due to environmental and other factors. Therefore, the processor can also update the state information stored in the historical data memory during a single discharge cycle. The data communication module can package the updated state information into a training dataset and send it to the cloud server, allowing the cloud server to retrain the future operating condition prediction model based on this updated training dataset. This enables the energy storage device provided in this application to have adaptive characteristics, thereby effectively improving the prediction accuracy of the energy state of the energy storage device.

[0009] Typically, the status information of a single battery cell includes temperature information. Therefore, the battery management system may also include a temperature signal converter, which can be used to convert the temperature information into parameters indicated by the processor, so that the processor can process it.

[0010] In addition, to enable the battery acquisition interface to collect temperature information from individual battery cells, the interface can be connected to the surface of the cell via a thermocouple. Specifically, the battery acquisition interface can be connected to the surface of the battery cell that is close to its electrode tab via a thermocouple, thereby improving the accuracy of the temperature information collected by the interface and thus enhancing the accuracy of energy state prediction for the energy storage device.

[0011] Secondly, this application provides an energy storage system, which may include a power converter and an energy storage device as described in the first aspect. The power converter can be used to convert the current and / or voltage input to the energy storage device, or the current and / or voltage output from the energy storage device, into power. In this energy storage system, cloud-based collaboration can achieve a more accurate prediction of the future operating condition information sequence of the energy storage device, thereby effectively improving the energy state of the energy storage device calculated based on the future operating condition information sequence, and ultimately obtaining the energy state of the energy storage system.

[0012] Thirdly, this application provides a method for predicting the operating conditions of an energy storage device, the energy storage device including a single battery cell, the method comprising:

[0013] Obtain the state information of individual battery cells during operation, and package the state information of individual battery cells from the start of discharge to the end of discharge into a training dataset;

[0014] Obtain the working condition prediction model trained based on the training dataset;

[0015] Based on the operating condition prediction model and the state information of individual battery cells, the future operating condition information sequence of the energy storage device is calculated.

[0016] The method provided in this application can predict the future operating condition information sequence of an energy storage device more accurately, thereby improving the accuracy of the energy state of the energy storage device calculated based on the predicted future operating condition information sequence.

[0017] Because the operating state of a single battery cell may fluctuate significantly due to factors such as the surrounding environment, some abnormal data may be collected in the state information during a discharge cycle from the start to the end of discharge. To avoid this abnormal data affecting the accuracy of predicting future operating condition information sequences, in one possible implementation of this application, the method further includes:

[0018] Abnormal data in the state information of individual battery cells during the process from the start to the end of discharge is removed. The remaining state information is then classified and preprocessed, and packaged to obtain the training dataset.

[0019] Since there are no outliers in this training dataset, using it to train a pre-established future working condition prediction model can yield more accurate prediction model parameters, thus making the resulting prediction model more accurate. This is beneficial for improving the accuracy of prediction results for future working condition information sequences.

[0020] As energy storage devices are used daily, the state information of individual battery cells can be affected by environmental factors or the age of the battery. Therefore, to ensure that the energy state of the energy storage device can still be predicted relatively accurately, the method described in this application may further include:

[0021] The training dataset is updated during a single discharge cycle, from the start to the end of the discharge.

[0022] In one possible implementation of this application, the above method may further include:

[0023] Acquire voltage, current, and temperature data of individual battery cells during constant current discharge testing under different conditions;

[0024] An initial state estimation model is established based on voltage, current, and temperature data.

[0025] The initial energy state of a single battery cell at the start of discharge is estimated based on the initial state estimation model.

[0026] This initial energy state can be used as an initial value for predicting the energy state of an energy storage device. Based on this initial value and combined with the operating state of the energy storage device, the future energy state of the energy storage device can be obtained.

[0027] In one possible implementation of this application, the establishment of the initial state estimation model based on voltage data, current data, and temperature data specifically includes:

[0028] Calculate the maximum usable energy released by a single battery cell under different conditions based on voltage and current data;

[0029] The initial state estimation model is obtained by fitting the maximum available energy.

[0030] Considering that as the battery discharges, the battery temperature gradually rises, while the internal resistance and polarization gradually decrease, some of the energy that could not be released due to polarization and internal resistance at low temperatures can now be released normally. Therefore, a temperature compensation factor is needed to compensate for and correct the energy state of individual battery cells, improving the accuracy of energy state estimation during the initial and early stages of discharge. Based on this, in one possible implementation of this application, the above method may further include:

[0031] Based on the pre-established temperature compensation model and the predicted future operating condition information sequence, the energy state compensation value of the energy storage device is obtained.

[0032] The prior energy state of the energy storage device is calculated based on the future operating condition information sequence and the state information of individual battery cells.

[0033] The energy state of the energy storage device is obtained based on the prior energy state and the energy state compensation value.

[0034] In this application, the process of establishing the temperature compensation model is not specifically limited. In one possible implementation of this application, the above method may further include:

[0035] Acquire temperature data of individual battery cells during constant current discharge testing under different conditions;

[0036] A temperature compensation model is obtained by fitting the discharge data of the same initial state of charge of the battery cells at different temperatures.

[0037] As can be seen from the above description of the energy storage device provided in this application, the processor can calculate the energy state of the energy storage device based on the future operating condition information sequence of the energy storage device. Based on this, an equivalent circuit model can be established in the processor. In specific implementations, the above method may further include:

[0038] The open-circuit voltage-state-of-charge curve of a battery cell is obtained by performing an open-circuit voltage test on the individual battery cell.

[0039] Based on the hybrid power pulse characteristic test of the battery cells, and the offline identification of the model parameters of the battery cells under different states of charge and temperature conditions using a genetic algorithm;

[0040] An equivalent circuit model of a single battery cell is established based on the open-circuit voltage-state-of-charge curve and model parameters.

[0041] The prior energy state of the energy storage device is obtained based on the equivalent circuit model and the sequence of future operating conditions.

[0042] Fourthly, this application provides a computer-readable storage medium having a computer program or instructions stored thereon, which, when executed, causes a computer to perform the method in any of the possible implementations of the third aspect described above.

[0043] Fifthly, this application provides a computer program product that, when executed by a computer, causes the computer to perform the method in any possible implementation of the third aspect described above. Attached Figure Description

[0044] Figure 1A schematic diagram of the structure of an energy storage device provided in an embodiment of this application;

[0045] Figure 2 Flowchart of the operating condition prediction method for an energy storage device provided in the embodiments of this application;

[0046] Figure 3 An open-circuit voltage-state-of-charge curve of an energy storage device provided in an embodiment of this application.

[0047] Figure label:

[0048] 1-Battery cell; 2-Battery management system; 201-Battery acquisition interface; 202-Processor; 203-Temperature signal converter;

[0049] 3-Working condition prediction and compensation device; 301-Historical data storage device; 302-Data communication module; 303-Compensation device;

[0050] 4-Shell; 5-Thermocouple. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. However, the exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein. The same reference numerals in the figures denote the same or similar structures, and therefore repeated descriptions of them will be omitted. The terms expressing position and direction described in the embodiments of this application are illustrative based on the accompanying drawings, but changes can be made as needed, and all such changes are included within the scope of protection of this application. The accompanying drawings of the embodiments of this application are for illustrating relative positional relationships only and do not represent actual scale.

[0052] It should be noted that specific details are set forth in the following description to facilitate understanding of this application. However, this application can be implemented in many ways other than those described herein, and those skilled in the art can make similar extensions without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0053] In the description of the embodiments of this application, "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The character " / " generally indicates that the related objects before and after it are in an "or" relationship. "At least one" in this application refers to one or more; "multiple" refers to two or more. Furthermore, it should be understood that in the description of this application, terms such as "first" and "second" are used only for descriptive purposes and should not be construed as indicating or implying relative importance or order.

[0054] To facilitate understanding of the energy storage device and its operating condition prediction method provided in this application embodiment, its application scenarios are first introduced below. An energy storage system is a device that stores electrical energy through a certain medium and releases the stored energy to generate electricity when needed. It can be used as a load balancing device and backup power supply in electronic devices such as servers and supercomputers, or in electric vehicles. Depending on the power consumption requirements of its application scenario, energy storage systems can be specifically divided into module-level energy storage systems, rack-level energy storage systems, and container-level energy storage systems. An energy storage system generally includes an energy storage device and a power converter. The energy storage device is the main device in the energy storage system that stores electrical energy. The power converter can be used to convert the current input to or output from the energy storage device into power, or it can also be used to convert the voltage input to or output from the energy storage device into power, so that the power of the energy storage system matches that of external power supply equipment or electrical equipment.

[0055] Taking energy storage systems used in electric vehicles as an example, to meet the energy demands of electric vehicles, the battery packs of energy storage systems are typically formed by connecting multiple energy storage devices in series. Each energy storage device contains multiple battery cells connected in series, and each battery cell consists of multiple cells connected in parallel. When connecting the various energy storage devices, copper or aluminum plates are generally used to connect adjacent energy storage devices, and screws are tightened to secure them together.

[0056] Lithium-ion batteries have become the preferred energy storage solution for electric vehicles and the energy storage industry due to their advantages such as high operating voltage, high specific energy, long cycle life, low self-discharge rate, and no memory effect. In the future, lithium-ion batteries will continue to trend towards larger sizes and modular designs. Based on this, the requirements for lithium-ion battery state of energy (SGE) monitoring are gradually increasing. The SGE of lithium-ion batteries is crucial for estimating the driving range and energy distribution of energy storage systems. Therefore, it is necessary to establish corresponding SGE prediction methods for lithium-ion batteries to accurately predict their energy levels, thereby providing a basis for the operation of the Battery Management System (BMS).

[0057] Current methods for predicting battery state of energy (SGE) typically involve either calculating the SGE using an equivalent circuit model or training a black-box model using offline data. The method using an equivalent circuit model requires first obtaining the parameters of the battery's equivalent circuit model through online parameter identification, and then using a Kalman filter to calculate the SGE. However, this method is only applicable to dynamic operating conditions; the results of parameter identification diverge under steady-state conditions. The method of training a black-box model using offline data requires a large amount of offline data to train the model. The accuracy of the final black-box model depends primarily on the completeness of the initial offline training data, but obtaining various operating condition data for a battery under different aging states is quite difficult.

[0058] The energy storage device operation condition prediction method provided in this application can be used to achieve relatively accurate prediction of the energy state of the energy storage device. The application will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0059] First, refer to Figure 1 , Figure 1 This is a schematic diagram of an energy storage device provided in an embodiment of this application. The energy storage device may include individual battery cells 1. The number of individual battery cells 1 in the energy storage device is not limited in this application. For example,... Figure 1 The energy storage device shown may include at least two battery cells 1, which may be connected in series, meaning that the current flowing through each battery cell 1 is the same.

[0060] Additionally, the energy storage device may also include a battery management system 2 and a condition prediction and compensation device 3. In this application, the specific locations of the battery management system 2 and the condition prediction and compensation device 3 within the energy storage device are not limited. For example, the energy storage device may also include a housing 4, which has a receiving cavity into which the battery cell 1 can be received, while the battery management system 2 and the condition prediction and compensation device 3 can be disposed on the housing 4. Figure 1 In the energy storage device shown, both the battery management system 2 and the operating condition prediction and compensation device 3 are located on the outside of the housing 4, so as to provide more space for the battery cells 1 and other modules of the energy storage device within the accommodating cavity. In some other possible embodiments of this application, when the space within the accommodating cavity is sufficient, the battery management system 2 and the operating condition prediction and compensation device 3 may also be located within the accommodating cavity.

[0061] When configuring the battery management system 2, you can continue to refer to... Figure 1The battery management system 2 may include a battery acquisition interface 201 and a processor 202. The battery acquisition interface 201 can be used to acquire the status information of the individual battery cells 1 during operation. It is worth noting that the status information of the individual battery cells 1 may include temperature information, current information, and voltage information, etc.

[0062] In addition, the status information collected by the battery acquisition interface 201 can be sent to the processor 202 so that the processor 202 can process the status information. In this application, the method by which the processor 202 processes the status information is not limited. Exemplary methods may include removing abnormal data from all status information of the battery cell 1 during the process from the start to the end of discharge, and performing classification preprocessing on the remaining status information.

[0063] Since the processor 202 cannot directly process temperature information under normal circumstances, a temperature signal converter 203 can be set in the battery management system 2. This temperature signal converter 203 can convert the temperature information into parameters indicated by the processor 202, thereby enabling the processor 202 to process the temperature information. The parameters indicated by the processor 202 can be, for example, voltage parameters.

[0064] As described above, the battery acquisition interface 201 can collect temperature information of the battery cell 1. In specific implementation, the battery acquisition interface 201 can collect the surface temperature of the battery cell 1, especially the temperature of the area near the electrode tab of the battery cell 1. This is because the temperature at this location is closer to the internal temperature of the battery cell 1, thus making the temperature information of the battery cell 1 collected by the battery acquisition interface 201 more accurate, which helps to improve the accuracy of predicting the energy state of the energy storage device.

[0065] In order for the battery acquisition interface 201 to acquire the surface temperature of the battery cell 1, please refer to... Figure 1 The battery acquisition interface 201 can be connected to the surface of the battery cell 1 via a thermocouple 5. In other possible embodiments of this application, the battery acquisition interface 201 can also be thermally connected to the surface of the battery cell 1 via other components with high thermal conductivity. These are not listed here, but they should all be understood to fall within the protection scope of this application.

[0066] You can continue to refer to Figure 1The operating condition prediction and compensation device 3 may include a historical data storage 301 and a data communication module 302. The processor 202 can store the status information collected by the battery acquisition interface 201 in the historical data storage 301. The data communication module 302 can package the status information of the battery cell 1 stored in the historical data storage 301 during the discharge process from start to end into a training dataset and upload it to a cloud server. It can also receive the future operating condition prediction model trained by the cloud server based on the training dataset.

[0067] Furthermore, the future operating condition prediction model received by the data communication module 302 can be downloaded to the processor 202, enabling the processor 202 to calculate the future operating condition information sequence of the energy storage device based on this prediction model and using the state information of the battery cell 1 collected by the battery acquisition interface 201 as input. Based on this, the processor 202 can also obtain the prior energy state of the energy storage device according to the future operating condition information sequence and the state information of the battery cell 1.

[0068] As the temperature of the energy storage device gradually rises from the start to the end of discharge, the internal resistance and polarization of the battery cell 1 gradually decrease. This allows some of the energy that was previously unable to be released due to polarization and internal resistance at low temperatures within the battery cell 1 to be released normally. Therefore, a compensation value can be added to the prior energy state of the energy storage device obtained based on the future operating condition information sequence and the state information of the battery cell 1 to improve the accuracy of the energy state prediction.

[0069] In specific implementation, you can continue to refer to Figure 1 The operating condition prediction and compensation device 3 may further include a compensation device 303, which can be used to determine the energy state compensation value of the energy storage device based on a pre-established temperature compensation model and the predicted future operating condition information of the energy storage device. Based on this, the processor 202 can superimpose the aforementioned prior energy state and the energy state compensation value to obtain the energy state of the energy storage device, which can effectively improve the accuracy of the energy state prediction of the energy storage device.

[0070] As can be seen from the above introduction of the energy storage device provided in this application, the energy state of the energy storage device can be calculated based on the future operating condition information sequence. In other words, obtaining the future operating condition information sequence of the energy storage device is the key to obtaining the energy state of the energy storage device. Based on this, the embodiments of this application also provide a method for predicting the operating conditions of an energy storage device. The structure of the energy storage device can be referred to the description in the above embodiments, and will not be repeated here. Figure 2 A flowchart of the operating condition prediction method for the energy storage device provided in the embodiments of this application is shown below. Figure 2As shown, the operating condition prediction method for this energy storage device may include the following steps:

[0071] Step S101: Obtain the state information of the battery cell when it is working, and package the state information of the battery cell from the start of discharge to the end of discharge into a training dataset.

[0072] Step S102: Obtain the working condition prediction model trained based on the training dataset;

[0073] Step S103: Calculate the future operating condition information sequence of the energy storage device based on the operating condition prediction model and the state information of the individual battery cells.

[0074] The method provided in this application embodiment can predict the future operating condition information sequence of an energy storage device more accurately, thereby improving the accuracy of calculating the energy state of the energy storage device based on the predicted future operating condition information sequence.

[0075] It is understandable that the operating state of a battery cell may fluctuate significantly due to factors such as the surrounding environment. Therefore, during a discharge cycle of a battery cell, from the start to the end of discharge, some abnormal data may be collected in the state information. To avoid this abnormal data affecting the accuracy of predicting future operating condition information sequences, the method described in this application may further include:

[0076] Abnormal data in the state information of individual battery cells during the process from the start to the end of discharge is removed. The remaining state information is then classified and preprocessed, and packaged to obtain the training dataset.

[0077] Since there are no outliers in this training dataset, using it to train a pre-established future working condition prediction model can yield more accurate prediction model parameters, thus making the resulting prediction model more accurate. This is beneficial for improving the accuracy of prediction results for future working condition information sequences.

[0078] Furthermore, in this application, the process of training the working condition prediction model based on the training dataset can be completed on a cloud server, and the specific training process can be as follows:

[0079] Obtain the training dataset and use the expectation-maximization algorithm to obtain the model parameters of the future working condition prediction model based on the Hidden Markov Model.

[0080] The trained operating condition prediction model is obtained based on the model parameters of the future operating condition prediction model and the pre-established future operating condition prediction model.

[0081] In practice, the first inputs are the observed variable data Y, the latent variable data Z, and the joint distribution. Conditional distribution ;

[0082] Then, select the initial values ​​for the model parameters. And begin iterating, the iterative steps of which can be:

[0083] Step 1: Remember For the first i Next iteration parameters The estimated value, in the first i In the +1st iteration, calculate:

[0084] ;

[0085] here, Given observation data Y and current model parameter estimation Hidden variable data Z The conditional probability distribution.

[0086] Step Two: Find the reason why Maximization ,Sure i Parameter estimates from +1 iterations :

[0087] ;

[0088] Finally, repeat steps one and two until convergence. The obtained model parameters... The initial state probability vector π, the state transition probability matrix A, and the observation probability matrix B can be represented as follows:

[0089] ;

[0090] ;

[0091] .

[0092] By inputting the above model parameters into the pre-established future operating condition prediction model, the trained operating condition prediction model can be obtained.

[0093] As energy storage devices are used daily, the state information of individual battery cells can be affected by environmental factors or the age of the battery. Therefore, to ensure that the energy state of the energy storage device can still be predicted relatively accurately, the method described in this application may further include:

[0094] The training dataset is updated during a single discharge cycle, from the start to the end of the discharge.

[0095] It is understandable that the prediction of the future energy state of an energy storage device is usually based on its initial energy state, combined with its operating state. Therefore, estimating the initial energy state of the energy storage device is a crucial step in achieving a more accurate prediction of its future energy state. Based on this, the above method may also include:

[0096] Acquire voltage, current, and temperature data of individual battery cells during constant current discharge testing under different conditions;

[0097] An initial state estimation model is established based on voltage, current, and temperature data.

[0098] The initial energy state of a single battery cell at the start of discharge is estimated based on the initial state estimation model.

[0099] In practice, firstly, constant current discharge tests can be conducted on individual battery cells under different temperatures, discharge rates, and initial states of charge (SOC) conditions, and voltage, current, and temperature data can be obtained during the discharge process.

[0100] Then, using the acquired voltage and current data, the maximum usable energy released by a single battery cell under different temperatures, discharge rates, and SOCs is calculated. Based on the calculation results, an initial state of energy prediction model for the energy storage device is fitted. .

[0101] Considering that the initial state of energy of an energy storage device is mainly affected by three factors—initial SOC, temperature, and discharge current—and that performing a complete calibration experiment would be very time-consuming, these factors are decoupled and implemented as follows: and .

[0102] Existing experiments have shown that, under the same temperature and discharge current conditions, the usable energy released by an energy storage device at different initial states of charge (SOC) exhibits a essentially linear relationship. Therefore, the calibration experiments described above only require testing at different temperatures and discharge currents with a small initial SOC. Thus, in this application, the initial energy state of the energy storage device... The calculation formula is as follows:

[0103]

[0104] in, These represent the initial SOC as follows: The calibration function obtained under the given conditions; This represents the initial temperature of the battery; This represents the initial discharge current of the energy storage device, or the discharge current of the energy storage device after a long period of rest. This represents the initial SOC of the battery.

[0105] In this way, based on the discharge data of the energy storage device at the same initial SOC but different temperatures, the temperature at the end of the discharge of the energy storage device can be used as the abscissa and the total energy released by the energy storage device can be used as the ordinate to fit a temperature correction curve and determine the correction coefficient K.

[0106] As can be seen from the above description of the energy storage device provided in this application, the processor can calculate the energy state of the energy storage device based on the future operating condition information sequence of the energy storage device. Based on this, an equivalent circuit model can be established in the processor. In specific implementations, the above method may further include:

[0107] The open-circuit voltage-state-of-charge curve of a battery cell is obtained by performing an open-circuit voltage test on the individual battery cell.

[0108] Based on the hybrid power pulse characteristic test of the battery cells, and the offline identification of the model parameters of the battery cells under different states of charge and temperature conditions using a genetic algorithm;

[0109] An equivalent circuit model of a single battery cell is established based on the open-circuit voltage-state-of-charge curve and model parameters.

[0110] The prior energy state of the energy storage device is obtained based on the equivalent circuit model and the sequence of future operating conditions.

[0111] The open-circuit voltage-state-of-charge curve can be seen as follows: Figure 3 As shown. Furthermore, offline identification of model parameters for individual battery cells under different states of charge and temperature conditions using a genetic algorithm may include offline identification of model parameters such as ohmic internal resistance, polarization internal resistance, and time constant. The equivalent circuit model of the battery cell established using the above method could be, for example:

[0112] ;

[0113] in, and These are the polarization internal resistance and polarization capacitance, respectively. This is the voltage drop of the RC parallel circuit, used to simulate the polarization voltage of a single battery cell; For current; This refers to the terminal voltage of a single battery cell; This is the open-circuit voltage of a single battery cell; This represents the ohmic internal resistance of a single battery cell.

[0114] Constructing the transfer function of a first-order RC model:

[0115] ;

[0116] make Then we have:

[0117] ;

[0118] Using the bilinear transformation method, the system's equations based on the s-plane are mapped to the z-plane; here, These are coefficients related to the model parameters;

[0119] ;

[0120] Transforming to the discrete domain, the result is:

[0121] ;

[0122] Due to the influence of the open-circuit voltage of a single battery cell on its state of charge (SOC), temperature (T), and aging condition... It has coupling properties, and defines the open-circuit voltage at time k. SOC value Temperature value and aging state function ,Right now:

[0123] ;

[0124] Assuming that the amount of electricity consumed or absorbed by a single battery cell has approximately zero impact on its State of Charge (SOC) per unit time, the temperature of the battery cell remains approximately constant, and the aging state of the battery cell remains approximately constant, then:

[0125] ;

[0126] In the discrete domain, we have:

[0127] ;

[0128] Then the RC transfer function in the discrete domain mentioned above can be reduced to:

[0129]

[0130] This allows us to obtain the equivalent circuit model of a single battery cell.

[0131] It is understood that in this application, by importing the state information of the acquired battery cells as input into the trained operating condition prediction model, the future operating condition sequence of the energy storage device can be obtained. Among these, the future current sequence of the energy storage device... for:

[0132] .

[0133] In addition, the predicted sequence of the state of charge (SOC) of a single battery cell can be calculated using the ampere-hour integration method. :

[0134] ;

[0135] ;

[0136] in This represents the prediction of the current sequence of a single battery cell over the next n seconds at second t=j. This represents the predicted SOC sequence of a single battery cell over the next n seconds at second t=j. CAP This represents the maximum capacity of a single battery cell.

[0137] By inputting the predicted current sequence into the equivalent circuit model, the predicted voltage sequence of the battery cell can be obtained. :

[0138]

[0139] The predicted current, voltage, and SOC sequences are input into the battery lumped-parameter thermal model to obtain the predicted temperature sequence. :

[0140] .

[0141] Considering that as the battery discharges, the battery temperature gradually rises, and at the same time the battery's internal resistance and polarization gradually decrease, this allows some of the energy that the battery could not release due to polarization and internal resistance at low temperatures to be released normally.

[0142] Therefore, a temperature compensation factor is needed to correct the energy state of individual battery cells, improving the accuracy of energy state estimation during the initial and early stages of discharge. In practice, the current temperature of the battery cell can be input based on the predicted temperature sequence. and the predicted temperature at the end of the discharge Combined with the calibrated temperature compensation coefficient k, the energy state compensation value DSOE of the energy state is calculated;

[0143]

[0144] Then, the prior energy state obtained above can be compensated based on the calculated energy state compensation value, wherein the prior energy state can be expressed as:

[0145]

[0146] The energy state obtained after compensating the prior energy state obtained above with the energy state compensation value is as follows:

[0147]

[0148] The method provided in this application primarily estimates the state of energy (SGE) of an energy storage device based on a Hidden Markov Model (HMM) and the equivalent circuit model of a single battery cell. It fully utilizes historical data from the individual battery cells within the energy storage device, combining it with an unsupervised learning algorithm to establish a predictive model for the device's operating conditions, thereby achieving accurate predictions of the device's future operating conditions. Furthermore, the predicted future operating condition sequence, combined with the equivalent circuit model, enables precise estimation of the energy SGE of the energy storage device. Additionally, this application allows for the storage and continuous updating of historical SGE information from individual battery cells. This continuously updated SGE information is used to retrain the predictive model, giving the method an adaptive characteristic. It enables rapid updates to the SGE algorithm through transfer learning when the operating conditions of the energy storage device change.

[0149] Based on the foregoing and the same concept, this application provides a computer-readable storage medium storing a computer program or instructions thereon, which, when executed, causes a computer to perform the methods described in the above-described method embodiments. The storage medium can be any available medium accessible to a computer. For example, but not limited to, a computer-readable medium may include RAM, read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), CD-ROM or other optical disk storage, magnetic disk storage media, or other magnetic storage devices, or any other medium capable of carrying or storing desired program code having an instruction or data structure form and accessible to a computer.

[0150] Based on the above content and the same concept, this application provides a computer program that, when run on a computer, causes the computer to perform the methods described in the above method embodiments.

[0151] It should be understood that the module division in the embodiments of this application is illustrative and only represents a logical functional division. In actual implementation, there may be other division methods. Furthermore, the functional modules in the various embodiments of this application can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0152] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the flow or function according to embodiments of this application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, an access network device, a terminal device, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. Computer-readable storage media can be any usable medium that a computer can access, or a data storage device such as a server or data center that integrates one or more usable media. Usable media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., digital video discs (DVDs)), or semiconductor media, etc.

[0153] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0154] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0155] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0156] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. An energy storage device, the energy storage device comprising a single battery cell, characterized in that, The energy storage device further includes a battery management system and a condition prediction and compensation device. The battery management system includes a battery acquisition interface and a processor. The condition prediction and compensation device includes a data communication module and a historical data storage device, wherein: The battery acquisition interface is used to acquire the status information of the individual battery cells during operation; The processor is used to store the status information into the historical data storage device; The data communication module is used to package the state information of the battery cell stored in the historical data storage device during the process from the start to the end of discharge into a training dataset and upload it to the cloud server; and is used to receive the future operating condition prediction model trained by the cloud server based on the training dataset. The processor is further configured to calculate a sequence of future operating conditions for the energy storage device based on the future operating condition prediction model and the state information of the individual battery cells. The processor is also configured to calculate the prior energy state of the energy storage device based on the future operating condition information sequence and the state information of the battery cells; The operating condition prediction and compensation device also includes a compensation device, which is used to determine the energy state compensation value of the energy storage device based on a pre-established temperature compensation model and in combination with the predicted future operating condition information sequence of the energy storage device. The processor obtains the energy state of the energy storage device based on the prior energy state and the energy state compensation value.

2. The energy storage device as described in claim 1, characterized in that, The processor is also used to update the status information stored in the historical data memory during a single discharge process, from the start of the discharge to the end of the discharge.

3. The energy storage device as described in claim 1 or 2, characterized in that, The status information includes temperature information, and the battery management system further includes a temperature signal converter, which is used to convert the temperature information into parameters indicated by the processor.

4. The energy storage device as described in claim 3, characterized in that, The battery acquisition interface is connected to the surface of the battery cell via a thermocouple.

5. An energy storage system, characterized in that, It includes a power converter and an energy storage device as described in any one of claims 1 to 4, wherein the power converter is used to perform power conversion on the current and / or voltage input to the energy storage device, or the current and / or voltage output from the energy storage device.

6. A method for predicting the operating conditions of an energy storage device, the energy storage device comprising individual battery cells, characterized in that, The method includes: Obtain the state information of the battery cell when it is working, and package the state information of the battery cell from the start of discharge to the end of discharge into a training dataset; Obtain the working condition prediction model trained based on the training dataset; Based on the operating condition prediction model and the state information of the individual battery cells, the future operating condition information sequence of the energy storage device is calculated. Based on the pre-established temperature compensation model and the predicted future operating condition information sequence, the energy state compensation value of the energy storage device is obtained; The prior energy state of the energy storage device is calculated based on the future operating condition information sequence and the state information of the individual battery cells. The energy state of the energy storage device is obtained based on the prior energy state and the energy state compensation value.

7. The method as described in claim 6, characterized in that, The method further includes: Abnormal data in the state information of the battery cell during the process from the start to the end of discharge is removed, the remaining state information is classified and preprocessed, and then packaged to obtain the training dataset.

8. The method as described in claim 6 or 7, characterized in that, The method further includes: The training dataset is updated during a single discharge process, from the start to the end of the discharge.

9. The method as described in claim 6 or 7, characterized in that, The method further includes: The voltage, current, and temperature data of the battery cells were obtained during the constant current discharge test under different conditions. An initial state estimation model is established based on the voltage data, the current data, and the temperature data; The initial energy state of the battery cell at the start of discharge is estimated based on the initial state estimation model.

10. The method as described in claim 9, characterized in that, The step of establishing the initial state estimation model based on the voltage data, the current data, and the temperature data includes: Calculate the maximum usable energy released by the battery cell under different conditions based on the voltage data and the current data; The initial state estimation model is obtained by fitting the maximum available energy.

11. The method as described in claim 6, characterized in that, The method further includes: The temperature data of the battery cell during constant current discharge testing under different conditions were obtained; The temperature compensation model is obtained by fitting the same initial state of charge of the battery cells and discharge data under different temperature conditions.

12. The method as described in claim 6 or 7, characterized in that, The method further includes: The open-circuit voltage-state-of-charge curve of the battery cell is obtained by performing an open-circuit voltage test on the battery cell. Based on the hybrid power pulse characteristic test of the battery cell, and the offline identification of the model parameters of the battery cell under different states of charge and temperature conditions using a genetic algorithm; An equivalent circuit model of the battery cell is established based on the open-circuit voltage-state-of-charge curve and the model parameters. The prior energy state of the energy storage device is obtained based on the equivalent circuit model and the future operating condition information sequence.

Images