Method for the optimized simulation of a characteristic of a battery cell

Abstract

The present invention relates to a method for the optimized simulation of a characteristic of a battery cell, containing a method for determining a varying internal resistance (RG) of a battery cell, in which method an electrochemical impedance spectroscopy is measured on the battery cell and an internal resistance (RG) of the battery cell is determined from a difference between a value of the locus curve with respect to a high-frequency measurement point (MPH) and a value of the locus curve with respect to a low-frequency measurement point (MPN) on the real-value axis (RE) of the Nyquist plot. In the further steps, a characteristic of the battery cell is simulated, comprising at least one output parameter from a simulated voltage (US), a simulated temperature (TS), a simulated state of charge (SOC) and a power limit, on the basis of a battery cell model (BM) according to input parameters comprising at least the previously measured current (I) and the previously measured temperature (TM). The simulated output parameter of the state of charge (SOC) is corrected on the basis of a predictive filter algorithm according to the measured voltage (UM) and the simulated voltage (US). Finally, to optimize the battery cell model (BM), a stored model parameter of the internal resistance (RG) is updated using a value which has been determined by the method for determining a varying internal resistance (RG) of a battery cell.

Description

[0001] PP34226WO / mg September 18, 2025 AVL List GmbH

[0002] Method for optimized simulation of a battery cell characteristic

[0003] The present invention relates to a method and a system for optimized simulation of a characteristic of a battery cell, which takes into account an internal resistance of the battery cell determined by a method for determining a varying internal resistance of a battery cell.

[0004] Model-based battery cell simulations are used, for example, in battery management systems (BMS) on board vehicles to determine the state of charge (SOC) of a battery storage system during operation. In today's battery storage systems, the state of charge of the battery cells is constantly monitored. This monitoring process, as well as the determination and control of various other parameters, is implemented by a battery management system to which the battery cells of the storage system are connected via sensors. A model-based simulation of the specific battery cells used in the storage system is used to determine the state of charge.

[0005] A suitable battery cell simulation model, which realistically depicts the properties and behavior of battery cells in actual battery storage operation under data processing from input parameters to output parameters, must first be developed, verified through characterization tests, and implemented in product development. The developed battery cell simulation model incorporates, among other things, resistance values ​​corresponding to an electrochemical characteristic of the battery cell, such as internal resistance (i.e., DC resistance). These values ​​are used as model parameters to simulate cell voltage, cell temperature, state of charge, and, if applicable, a power limit, etc. In particular, the cell voltage simulated by the battery cell simulation model is used by a further algorithm to estimate the state of charge.

[0006] In the current state of the art, the resistance parameters of a battery cell required for the battery cell simulation model are first determined in a laboratory environment using complex electrochemical impedance spectroscopy measurements (PP34226WO / mg 18.09.2025 AVL List GmbH) to develop the battery cell simulation model. At a later stage, a correlation between the battery cell simulation model and real-world operation can be verified by comparing simulated and measured voltages. If necessary, model parameters such as resistance parameters are adjusted or optimized in the laboratory environment down to a predefined error tolerance. Subsequently, the battery cell simulation model established under laboratory conditions is added to the battery management system of an application system and used, for example, throughout the entire lifecycle of a battery storage system.

[0007] However, the resistance parameters of the battery cells vary depending on other operating parameters, such as the state of charge, temperature changes, and the ongoing aging of the battery cells. Thus, the resistance parameters, originally determined in the laboratory, change both over the life cycle of the battery cells and under fluctuating operating states and conditions. Therefore, determinations of state of charge based on constant model parameters are associated with a higher error rate.

[0008] Looking at another application, model-based battery cell simulations are also used, for example, in the cell characterization process of battery cells with new specifications during the development and integration phase for a specific product application. In this context, usage-related performance limits must be determined, which relate to the thermally permissible effects of charging or discharging with a direct current due to the internal resistance of the battery cell. Here, too, the performance limits depend on other parameters and can vary with the state of charge or temperature.

[0009] In another application example, a battery manufacturer needs to determine test-related performance limits for novel battery cells with unknown properties during characterization tests to identify and prevent damage. Battery cell manufacturers currently use various characterization test methods to determine the required parameters of battery cells for a battery cell simulation model (PP34226WO / mg 18.09.2025 AVL List GmbH). One of these is the Hybrid Pulse Power Characterization (HPPC) test. Before applying the HPPC test, the manufacturer should define various current levels to be applied under different temperature and state-of-charge conditions. However, a performance limit, in the sense of a current limit, for newly developed battery cells can only be determined after time-consuming testing.If the HPPC test is to be applied to a battery cell whose current limits are unknown, the battery cell may be damaged as a result of an excessively high current, or the test may not be time-efficient or not be measurably significant as a result of an excessively low current.

[0010] The prior art, as outlined in US patent 10,338,146 B2, discloses the use of a filter algorithm, such as a Kalman filter, between a simulated voltage and a measured voltage to improve the accuracy of estimations. However, this technique does not address the root cause of deviations in a model-based simulation and therefore does not allow for modification or continuous adjustment of the underlying parameter model. Consequently, changing properties of battery cells due to aging must be incorporated into further assessments using algorithms.

[0011] US patent 10,386,422 B2 discloses a real-time measurement technique for electrochemical impedance spectroscopy (EIG) during the operation of a battery cell system. A model of the battery cells is parameterized based on these real-time measurements, with an impedance response simulated by the model being fitted to the measured data from a Bode plot of the real-time measurements. The specific details of how and which parameters are used to fit the model to the real-time measurements are not disclosed. Since real-time EIG measurements during operation inherently involve significant measurement noise, it can be assumed that this technique introduces inaccuracies in the model parameters and requires substantial data processing for real-time execution.

[0012] It is an object of the invention to at least partially overcome the aforementioned disadvantages in the prior art. It is also an object of the invention to achieve a PP34226WO / mg 18.09.2025 AVL List GmbH

[0013] The invention aims to create a technique that enables a rapid or simplified determination of the varying internal resistance in a real battery cell. A further objective is to allow the consideration of this varying internal resistance in model parameters for simulating battery properties with limited technical effort, making it feasible with limited resources in a mobile application. An alternative objective is to reduce the resource expenditure for the initial or updated parameterization of a simulation model with respect to the internal resistance of a battery cell.

[0014] One, and in particular all, of the preceding problems are solved by a method for the optimized simulation of a characteristic of a battery cell comprising a method for determining a varying internal resistance of a battery cell with the steps of claim 1. Further features and details of the invention will become apparent from the dependent claims, the description and the drawings.

[0015] Features and details described in connection with the method according to the invention naturally also apply in connection with steps or device features of the method or system for the optimized simulation of a property of the battery cell and vice versa, so that mutual reference can always be made to the disclosure with regard to the individual aspects of the invention.

[0016] The procedure for determining a varying internal resistance of a battery cell comprises the following steps:

[0017] - Measuring electrochemical impedance spectroscopy on the battery cell with the following intermediate steps:

[0018] Applying measuring pulses of an alternating current to the potential terminals of the battery cell for a high-frequency measuring point and for a low-frequency measuring point, which differ with respect to the frequency of the respective alternating current, using measuring electrodes; and PP34226WO / mg 18.09.2025 AVL List GmbH

[0019] Capturing impulse responses resulting from the measurement pulses at the potential poles of the battery cell to the high-frequency measurement point and to the low-frequency measurement point using measuring sensors;

[0020] - Determining the internal resistance of the battery cell using the following intermediate steps:

[0021] - Determining the values ​​of a locus curve in a Nyquist diagram, which includes a real axis and an imaginary axis, from a frequency response of the recorded impulse responses to the high-frequency measurement point and to the low-frequency measurement point; and

[0022] - Determining the internal resistance of the battery cell from a difference between a value of the locus curve at the high-frequency measurement point and a value of the locus curve at the low-frequency measurement point on the real value axis of the Nyquist diagram.

[0023] Building upon this, the inventive method serves for the optimized simulation of a battery cell characteristic, comprising the following steps: providing a parameterized battery cell model that has several model parameters, including at least one resistance parameter, relating to properties of a battery cell represented by the battery cell model; measuring a current and a temperature relative to the battery cell; simulating the battery cell characteristic, comprising at least one output parameter from a simulated voltage, a simulated temperature, a simulated state of charge, and a power limit, based on the battery cell model as a function of input parameters, including at least the previously measured current and temperature; measuring a voltage of the battery cell;and include correcting the simulated state-of-charge output parameter based on a predictive filter algorithm depending on the measured and simulated voltages, and optimizing the battery cell model by updating a stored model parameter of internal resistance with a value determined by the method for determining a varying internal resistance of a battery cell. Thus, the updated model parameters are fed into an optimized simulation that achieves more accurate results. PP34226WO / mg 18.09.2025 AVL List GmbH;

[0024] The steps of measuring the voltage of the battery cell and correcting the simulated output parameter of the state of charge based on a predictive filter algorithm depending on the measured voltage and the simulated voltage improve the accuracy of the simulated state of charge.

[0025] The invention thus provides, for the first time, an analytically simplified method for determining the varying property of the internal resistance in a real battery cell.

[0026] According to the present disclosure, the following terms are defined in the context of the invention as follows:

[0027] - The internal resistance is an internal ohmic resistance, i.e., a DC resistance, which represents a kinetic inhibition of electron mobility and directly influences the output voltage of the battery cell, as well as being the cause of waste heat in the case of an electron flow from the active material via the potential poles, such as a discharge current from the battery cell, or in the opposite direction in the case of a charging current.

[0028] - Impedance is a resistance parameter that represents an internal AC resistance and is in relation to capacitance, i.e., a charge or amount of mobile electrons available to supply a load.

[0029] - The charge displacement resistance is a resistance parameter that represents a kinetic inhibition of ion mobility in the electrolyte and a charge transfer at a phase boundary between electrolyte and electrode.

[0030] - Diffusion resistance is a resistance parameter that reflects the distribution of freely available charge carriers such as lithium ions in the electrolyte, i.e., a state that can be affected by, for example, an immobilizing agglomeration of ions.

[0031] The resistance parameters of the impedance, the

[0032] Charge displacement resistance and diffusion resistance influence the internal resistance of the battery cell, as can be derived from a known equivalent circuit model. PP34226WO / mg 18.09.2025 AVL List GmbH

[0033] A Nyquist plot, which represents the frequency response of a measured impulse response from electrochemical impedance spectroscopy in a Nyquist diagram, can be roughly divided into three sections corresponding to different frequency ranges. These three Nyquist plot sections are characteristic for determining the three resistance parameters mentioned: impedance, charge displacement resistance, and diffusion resistance. For example, the values ​​of a Nyquist plot section corresponding to a frequency range of 500 Hz and above, preferably 1 kHz or above, are related to the actual value of the internal impedance of the battery cell. The values ​​of another Nyquist plot section, corresponding to a frequency range of 5 Hz to 500 Hz, preferably 10 Hz to 100 Hz, are related to the actual value of the charge displacement resistance of the battery cell.Again, values ​​of a further locus curve section, which is assigned to a frequency range of at most 5 Hz or less, preferably about 1 Hz or less, are in relation to the actual value of the diffusion resistance of the battery cell.

[0034] - The high-frequency measurement point of the electrochemical impedance spectroscopy measured according to the invention lies in a frequency range of 10 mHz to 10 Hz, and the low-frequency measurement point in a frequency range of 10 kHz to 1 MHz.

[0035] Building on this, a technical application of battery cell simulation using a battery cell model is provided, which benefits from the integration of the inventive method. Particularly in mobile applications with limited resources regarding measurement technology and computing power, the simulation result can now be optimized by considering changes in the actual internal resistance within the framework of updates to the input parameter of internal resistance underlying the battery cell model. This is only made possible by the analytically simplified method in autonomous operation according to the invention. PP34226WO / mg 18.09.2025 AVL List GmbH. This enables mobile systems such as...Electric vehicles with existing onboard measurement and data processing equipment are enabled to perform improved simulation-based state-of-charge determination of the battery cell even over advanced stages of the battery cell's life cycle, which, for example, allows for a more accurate determination of the vehicle's range.

[0036] The varying real internal resistance is determined for the first time using a simplified measurement and analytical implementation of electrochemical impedance spectroscopy (EIS) according to the invention. Consideration of the varying real internal resistance as an input parameter in the battery cell model for a realistic simulation of a battery cell characteristic, such as a simulated voltage or state of charge (SOC), is achieved by continuously determining the internal resistance in this simplified manner to update model parameters, preferably at time intervals or depending on parameter ratios.

[0037] The simplified determination of the battery cell's internal resistance according to the invention thus enables parameter optimization of the internal resistance under the input parameters of a simulation autonomously within an application system. This eliminates the need for the same application system to estimate the effects of aging on the parameters using a further algorithm, for example, in an external device. Furthermore, the accuracy of the simulated voltage and state of charge in a battery management system is improved throughout the battery's entire lifespan. An accurate assessment of the state of charge of a battery storage system is essential for many operating situations, such as charging processes or the aforementioned range estimation in a vehicle with a traction battery.

[0038] Electrochemical impedance spectroscopy is performed using a sensor network of measuring electrodes and voltage sensors on battery cells, which is already present in a conventional battery management system of a battery storage system. Instead of a conventional measurement over a comprehensive frequency spectrum, as is usually carried out in a laboratory environment, only a few, namely just two, measurement points are required for electrochemical impedance measurement at a high frequency and a low frequency. According to the findings and an analytical approach underlying the invention, the internal resistance values ​​of the battery cells are determined based on a map or frequency response of the measurement points in a Nyquist diagram, more precisely based on the difference between values ​​of a locus or function on the real value axis of the Nyquist diagram at the measurement points.

[0039] One advantage of the invention is that the technique required for this simplified measurement and analysis can be performed in a self-contained system, such as using existing components of a vehicle's electrical system or a battery management system (BMS). This allows the measurement and updating of the model parameters within the system or electrical system to be carried out during a rest phase, i.e., a period when no load is applied to a battery cell. This eliminates the need for an external update, an external resource, or recharacterization in a laboratory environment to update the parameterization of the battery cell model.

[0040] A further advantage of the invention is that the simplified and repeatedly performed internal resistance determination according to the invention enables individual and autonomous parameter acquisition of the internal resistance in the application product, thus creating, in principle, an opportunity for data pooling from the multitude of products and users used throughout the entire life cycle of the battery cells in a real-world environment. From a technical perspective, this data pooling requires only the transmission or retrieval of the individual measurement data or a history of the optimized model parameters into a central data storage system. The data collected in this way offers manufacturers significant benefits for potential analyses and insights.

[0041] In other applications by battery cell manufacturers or application system manufacturers where battery management of new battery cells needs to be implemented, the parameter optimization according to the invention, with reduced measurement and analysis effort, significantly reduces the testing effort of characterization tests on battery cells with indeterminate characteristics. PP34226WO / mg 18.09.2025 AVL List GmbH

[0042] Parameter generation. By creating a suitable battery cell model that covers the characteristics of the battery cell across comprehensive or application-specific areas, measurement and analytical resources, as well as the time and costs for test series on real battery cells, can be significantly reduced compared to conventional test procedures.

[0043] The performance limits that battery cells can provide based on their internal resistance under various load conditions depend on dynamic parameters such as the internal resistance, cell temperature, load, and / or state of charge. Depending on the dynamic nature of these parameters, the method according to the invention can be implemented in real-time test runs and as a software module, i.e., a parameter optimization module, or integrated into a battery management system as a simulation-based state-of-charge estimation module that compensates for the increasing internal resistance of the aging battery cell.

[0044] According to one aspect of the invention, the method for determining a varying internal resistance of a battery cell can comprise the additional step of: determining at least one structure-specific resistance parameter, comprising an impedance, a charge displacement resistance, and a diffusion resistance of the battery cell, from at least one of the following substeps: determining the impedance of the battery cell from a value or range of values ​​of the locus curve in the Nyquist diagram, relating to a frequency of at least 500 Hz or more, preferably about 1 kHz; determining the charge displacement resistance of the battery cell from a value or range of values ​​of the locus curve in the Nyquist diagram, relating to a frequency of at least 5 Hz to 500 Hz, preferably about 100 Hz;and / or determining the diffusion resistance of the battery cell from a value or range of values ​​of the locus curve in the Nyquist diagram, relating to a frequency of at most 5 Hz or less, preferably about 1 Hz or less.

[0045] This simplified determination approach allows further structure-related resistance parameters of the battery cell to be determined in real time by an algorithm and updated as stored input parameters of a simulation model PP34226WO / mg 18.09.2025 AVL List GmbH. If the measurement is performed in a standby state of the battery cell's application system, interference with the measurement can be minimized.

[0046] According to one aspect of the invention, the method for determining a varying internal resistance of a battery cell can be repeated at a predetermined time interval. This provides an alternative, simple solution for updating and optimizing the model parameters, the reduced effort required for measurement and updates taking into account the slow aging process of the battery cell.

[0047] According to one aspect of the invention, the method for determining a varying internal resistance of a battery cell can also include transmitting at least one previously determined resistance parameter from the internal resistance, impedance, charge displacement resistance, and diffusion resistance via a data interface to an external database. This makes evaluated measurement data and insights from individual applications available for data pooling.

[0048] According to one aspect of the invention, the method for optimizing the simulation of a battery cell characteristic can include monitoring changes in the measured temperature and changes in the simulated or corrected state of charge; and issuing a measurement request for optimizing the battery cell model, depending on the condition that the monitored change in the measured temperature exceeds a temperature change threshold, or that the monitored change in the simulated or corrected state of charge exceeds a charge change threshold. Thus, a parameter-based decision can be made regarding the need for a new determination of the model parameters, which further reduces the total number and cumulative effort required for measuring and updating the internal resistance over the battery's life cycle.

[0049] Based on this, a system for the optimized simulation of a battery cell can be divided into: a simulation module in which the parameterized battery cell model is provided; at least one data input interface for input parameters from a voltage, a current and a temperature, which were previously measured with respect to at least one battery cell; at least one data output interface for output parameters of a simulated state of charge, a simulated voltage, a temperature; and a parameter optimization module for optimizing the parameterized battery cell model by carrying out the inventive method for determining the internal resistance of the battery cell and by updating a stored model parameter to the internal resistance; wherein the data output interface of the parameter optimization module is connected to a data input interface of the simulation module.

[0050] According to one aspect of the invention, the said system for optimized simulation of a battery cell can be implemented in a battery management system of a battery storage system with multiple battery cells and utilize a network of sensors of the same.

[0051] Further advantages, features, and details of the invention will become apparent from the following description, in which exemplary embodiments of the invention are described in detail with reference to the drawings. The features and embodiments mentioned in the claims and in the description can each be essential to the invention individually or in any combination. The drawings schematically illustrate:

[0052] Fig. 1 shows a Nyquist diagram in which an analysis of two measurement points of an electrochemical impedance spectroscopy for the determination of the internal resistance of the battery cell according to the invention is shown;

[0053] Fig. 2 shows a representation of incoming and outgoing parameters in a software-based system for optimized simulation of a battery cell as an application environment of the invention; and

[0054] Fig. 3 shows a block diagram of the setup of the system from Fig. 2.

[0055] Fig. 1 shows a Nyquist plot with a real axis RE and an imaginary axis IM, in which measurement data from electrochemical impedance spectroscopy are plotted. The two circled measurements represent a high-frequency measurement point MPH and a low-frequency measurement point MPN, which were previously determined on the battery cell. For this purpose, measurement pulses of an alternating current at a higher and a lower frequency were applied to the potential terminals of the battery cell using measuring electrodes. Furthermore, an impulse response to the high-frequency measurement point MPH and the low-frequency measurement point MPN was recorded using voltage and current sensors. A frequency response of the recorded impulse responses was then converted for the calculation and plotting of the depicted locus curve in the Nyquist plot.The difference between the values ​​on the real-value axis RE of the high-frequency measurement point MPH and the low-frequency point MPN on the locus curve yields, in an analytically simplified manner, the internal resistance RG of the battery cell. This value can be determined by a real-time optimization algorithm as the new internal resistance RG of the battery cell and subsequently updated and stored as the relevant resistance parameter in a battery cell model BM for simulating a property in a battery management system or similar applications.

[0056] Figure 2 shows a schematic view of an embodiment of a software-based system 10 for simulating a characteristic, i.e., electrochemical or physical properties such as the voltage or state of charge of a battery cell, or of a plurality of battery cells in a battery storage system. The system 10 can, for example, be part of a battery management system (BMS) of a battery storage system, such as a traction battery in a vehicle, or part of a test setup in a manufacturer's laboratory environment.The system 10 comprises or has access to data processing means not shown, such as a CPU, to perform the determination of the internal resistance according to the invention and subsequent parameter optimization on a battery cell model, as well as a data input interface and a data output interface, the latter of which may be combined in a common data input and output interface.

[0057] Measurement data from electrochemical impedance spectroscopy (EIS) measurements, relating to a high-frequency measurement point MPH and a low-frequency measurement point MPN, are read in via the data input interface. An analysis of this measurement data will be described in more detail later. The measurements are performed by means of measurement technology PP34226WO / mg 18.09.2025 AVL List GmbH, which is either part of System 10 or accessible to System 10.These devices, which may already be an existing component of a conventional battery management system in a battery storage system, include a pulse generator capable of generating measurement pulses of alternating current at varying frequencies, applied to a battery cell, and sensors such as voltage and current sensors for individual battery cells, particularly suitable voltage sensors for high temporal resolution of measurement data, which measure a pulse response at the terminals of a single battery cell or a multitude of battery cells. The measured pulse response is a reaction of the chemical and physical structure of each battery cell to the measurement pulse.

[0058] In addition, to simulate a characteristic of a battery cell, the system 10 uses the data input interface to obtain a measured voltage UM at the poles of the real battery cell, i.e., a cell voltage detected by a voltage sensor on the battery cell, a measured current through the poles of the battery cell I detected by a current sensor, and a measured temperature TM measured on the battery cell or in a battery storage system with multiple battery cells by a temperature sensor.

[0059] As a result of the simulation, the system outputs various parameters via the data output interface, depending on the configuration for an application. These parameters can include a simulated state of charge (SOC) of the battery cell, determined by an algorithm and, if necessary, corrected by a filter; a measurement request (MA) issued under a condition described later for re-performing a measurement and updating parameters; a simulated voltage (US) of the battery cell; a simulated temperature (TS) of the battery cell; or an internal resistance (RG) determined from repeated measurements, i.e., a DC resistance of the battery cell.

[0060] The block diagram in Fig. 3 shows a more detailed structure of the software-based system 10, divided into functional modules. In particular, the system 10 comprises a parameter optimization module 11 for optimizing a parameterized battery cell model BM, a simulation module 12 with the parameterized battery cell model BM for simulating a characteristic of a battery cell, an optional filter module 13 for executing a predictive, corrective filter algorithm, such as a Kalman filter, and a measurement request module 14 for making a parameter-based decision on the necessity of remeasuring and updating model parameters and for issuing a request for a new measurement.

[0061] Initially, two new EIS measurements are performed on the battery cell in response to a measurement request MA; more precisely, a high-frequency impedance measurement and a low-frequency impedance measurement. The acquired measurement data from these two measurement points at different frequencies are referred to here as the high-frequency measurement point MPH and the low-frequency measurement point MPN.

[0062] The acquired measurement data are transferred to parameter optimization module 11, which analyzes, in particular, the axis projections of the high-frequency and low-frequency impedance measurements onto the real-value axis of a Nyquist diagram. This process is performed in a real-time optimization algorithm based on the understanding and approach that a value of the internal resistance RG, i.e., the DC resistance of the battery cell, correlates with a difference between the axis projections of the high-frequency measurement point MPH and the low-frequency measurement point MPN on the real-value axis. The real-time optimization algorithm further includes a cost function between the internal resistance RG value thus calculated and model parameters, in particular other resistance parameters, i.e., the impedance RW, a charge displacement resistance RV, and a charge diffusion resistance RD.The real-time optimization algorithm determines the impedance RW based on the difference between the axis projection of the high-frequency measurement point MPH and the low-frequency point MPN on the real value axis RE.

[0063] The resistance parameters determined from the measurement data, such as a parameter set consisting of the impedance RW, the charge displacement resistance RV, and the charge diffusion resistance RD, or the internal resistance RG determined in this context, are transferred to simulation module 12 and updated as stored values ​​in the battery cell model BM. Thus, the battery cell model BM, on which the simulation of the battery cell characteristics is based, has optimized model parameters that are adapted and updated to reflect any change in a property of the real battery cell determined through measurement, such as an age-related increase in internal resistance.

[0064] Simulation module 12 also takes measured values ​​of the current I through the battery cell and the temperature at the battery cell or in a battery storage system with multiple battery cells as input parameters for the simulation. The battery cell model BM is an equivalent circuit model that is continuously updated with the obtained resistance parameters RW, RV, RD, or RG, after which, as a result of the simulation, a state of charge (SOC), a simulated cell voltage (US), and a simulated temperature (TS) are reproduced at each iteration of the simulation.

[0065] The approach of updating the model parameters in real time at specific time intervals or upon the occurrence of parameter-based conditions using simplified electrochemical impedance measurements at only two measuring points, as described in the invention, limits or avoids the error rate that occurs in a conventional estimation of the voltage response, which increases over time due to the rise in resistance values ​​with the aging of the battery cells and which is conventionally performed by an algorithm without metrological verification. The more the resistance values ​​increase due to cell aging, the more this is reflected in the value of the internal resistance RG.However, this is determined by the parameter optimization module 11 through the described analysis of measurement data from the repeatedly performed and simplified electrochemical impedance measurement, stored in the battery cell model by updated values, and thus contributes to achieving optimized simulation results.

[0066] The simulation result of simulation module 12 includes, among other things, the output parameter of the simulated cell voltage US, which, in the illustrated embodiment of system 10, is inputted to an input of the predictive and corrective filter algorithm of filter module 13. The value of the actually measured cell voltage UM is input to another input of filter module 13. Based on the error difference between the value of the simulated voltage US and the value of the measured voltage UM, filter module 13 corrects the value of the simulated new state of charge (SOC) by increasing or decreasing the previous SOC value.

[0067] Electrochemical impedance spectroscopy, which conventionally encompasses a frequency spectrum, is a time-consuming measurement method that is technically impossible to perform continuously or in real time. Therefore, the parameter optimization module 11 performs only a limited number of measurements at just two or a few measurement points at cell-specific characteristic frequencies where relevant parameters are known to change. This prevents a rapid decrease in the battery cell's state of charge (SOC) during measurement and limits data processing effort. A change in the state of charge (SOC) by a specific percentage or a change in the simulated temperature (TS) by a specific amount serves as a suitable condition for verification, i.e., for triggering a repeat measurement.

[0068] Following this principle, the measurement request module 14 issues the measurement request MA to trigger new electrochemical impedance spectroscopy measurements. Beforehand, the measurement request module 14 checks whether the changes in the specified values ​​exceed a respective change threshold and then decides whether a new measurement should be performed as a prerequisite for optimizing and updating the model parameters. In this way, the change in impedance RW caused by the state of charge (SOC) or the temperature TS is always detected, and the increase in internal resistance RG in the real battery cell due to aging is automatically compensated for via parameter optimization and updating in the simulation.

[0069] The preceding descriptions of the embodiments illustrate the present invention solely by way of example. Naturally, individual features of the embodiments can be freely combined with one another, provided this is technically feasible, without departing from the scope of the present invention. PP34226WO / mg 18.09.2025 AVL List GmbH

[0070] Reference symbol list

[0071] 10 software-based systems

[0072] 11 Parameter optimization module

[0073] 12 Simulation module

[0074] 13 Filter module

[0075] 14 Measurement Request Module

[0076] BM battery cell model

[0077] I Electricity

[0078] IM Imaginary value axis

[0079] MA measurement requirement

[0080] MPN Low Frequency Measurement Point

[0081] MPH high-frequency measurement point

[0082] RD Diffusion resistance

[0083] RE Real Value Axis

[0084] RG internal resistance

[0085] RV charge displacement resistance

[0086] RW impedance

[0087] SOC charge level

[0088] TM measured temperature

[0089] TS simulated temperature

[0090] UM measured voltage

[0091] US simulated voltage

Claims

PP34226WO / mg September 18, 2025 AVL List GmbH Patent claims 1. Method for optimized simulation of a characteristic of a battery cell comprising a method for determining a varying internal resistance (RG) of a battery cell with the following steps: Measuring electrochemical impedance spectroscopy on the battery cell with the following intermediate steps: - Applying alternating current measurement pulses to the potential terminals of the battery cell for a high-frequency measurement point (MPH) and for a low-frequency measurement point (MPN), which differ with respect to the frequency of the respective alternating current, using measuring electrodes; and - Recording impulse responses resulting from the measurement pulses at the potential terminals of the battery cell to the high-frequency measurement point (MPH) and to the low-frequency measurement point (MPN) using measuring sensors; Determining the internal resistance (RG) of the battery cell using the following intermediate steps: - Determining the values ​​of a locus curve in a Nyquist diagram, which includes a real axis and an imaginary axis, from a frequency response of the recorded impulse responses to the high-frequency measurement point (MPH) and to the low-frequency measurement point (MPN); and - Determining the internal resistance (RG) of the battery cell from a difference between a value of the locus curve and the high-frequency measurement point (MPH) and a value of the locus curve to the low-frequency measurement point (MPN) on the real value axis (RE) of the Nyquist diagram; characterized by the further steps: Providing a parameterized battery cell model (BM) that includes several model parameters, including the internal resistance (RG) PP34226WO / mg 18.09.2025 AVL List GmbH includes, for the simulation of properties of a battery cell represented by the battery cell model (BM); Measuring a current (I) and a temperature (TM) in relation to the battery cell; Simulating the characteristics of the battery cell, comprising at least one output parameter from a simulated voltage (US), a simulated temperature (TS), a simulated state of charge (SOC) and a power limit, based on the battery cell model (BM) as a function of input parameters, comprising at least the previously measured current (I) and the previously measured temperature (TM); Measuring the voltage (UM) of the battery cell; Correcting the simulated state of charge (SOC) output parameter based on a predictive filter algorithm depending on the measured voltage (UM) and the simulated voltage (US); and Optimizing the battery cell model (BM) by updating a stored model parameter of the internal resistance (RG) with a value determined by the method for determining a varying internal resistance (RG) of a battery cell.

2. The method according to claim 1, characterized in that the method for determining a varying internal resistance (RG) of a battery cell comprises the additional step: Determining at least one further structure-specific resistance parameter, comprising an impedance (RW), a charge displacement resistance (RV) and a diffusion resistance (RD) of the battery cell, from at least one of the following substeps: - Determining the impedance (RW) of the battery cell from a value or range of values ​​of the locus curve in the Nyquist diagram, which relates to a frequency of at least 500 Hz or more, preferably about 1 kHz; PP34226WO / mg September 18, 2025 AVL List GmbH - Determining the charge displacement resistance (RV) of the battery cell from a value or range of values ​​of the locus curve in the Nyquist diagram, relating to a frequency of at least 5 Hz to 500 Hz, preferably about 100 Hz; and / or - Determining the diffusion resistance (RD) of the battery cell from a value or range of values ​​of the locus curve in the Nyquist diagram, relating to a frequency of at most 5 Hz or less, preferably about 1 Hz or less.

3. Method according to claim 1 or 2, characterized in that in the method for determining a varying internal resistance (RG) of a battery cell, the steps of measuring and determining are repeated depending on a predetermined time interval.

4. Method according to one of claims 1 to 3, characterized in that the method for determining a varying internal resistance (RG) of a battery cell comprises the additional step: - Transfer of at least one previously determined resistance parameter from the internal resistance (RG), the impedance (RW), the charge displacement resistance (RV) and the diffusion resistance (RD) via a data interface to an external database.

5. Method for optimized simulation of a characteristic of a battery cell according to one of claims 1 to 4, characterized by the steps: Monitoring changes in measured temperature (TM) and simulated or corrected state of charge (SOC); and Outputting a measurement request (MA) to optimize the battery cell model (BM) depending on a condition that the monitored change in the measured temperature (TM) exceeds a temperature change threshold, or the monitored change in the simulated or corrected state of charge (SOC) exceeds a charge change threshold, and PP34226WO / mg September 18, 2025 AVL List GmbH Electrochemical impedance spectroscopy measurements on the battery cell.

6. System (10) for optimized simulation of a battery cell characteristic by carrying out the method steps according to any one of claims 1 to 5, characterized by: a simulation module (12) in which the parameterized battery cell model (BM) is provided; at least one data input interface for input parameters from a voltage (UM), a current (I), and a temperature (TM) previously measured with respect to at least one battery cell; at least one data output interface for output parameters of a simulated state of charge (SOC), a simulated voltage (US), and a temperature (TS); and a parameter optimization module (11) for optimizing the parameterized battery cell model (BM) by carrying out a method for determining a varying internal resistance (RG) of a battery cell comprising the following steps: Measuring electrochemical impedance spectroscopy on the battery cell with the following intermediate steps: - Applying alternating current measurement pulses to the potential terminals of the battery cell for a high-frequency measurement point (MPH) and for a low-frequency measurement point (MPN), which differ with respect to the frequency of the respective alternating current, using measuring electrodes; and - Recording impulse responses resulting from the measurement pulses at the potential terminals of the battery cell to the high-frequency measurement point (MPH) and to the low-frequency measurement point (MPN) using measuring sensors; Determining the internal resistance (RG) of the battery cell using the following intermediate steps: PP34226WO / mg September 18, 2025 AVL List GmbH - Determining the values ​​of a locus curve in a Nyquist diagram, which includes a real axis and an imaginary axis, from a frequency response of the recorded impulse responses to the high-frequency measurement point (MPH) and to the low-frequency measurement point (MPN); and - Determining the internal resistance (RG) of the battery cell from a difference between a value of the locus curve at the high-frequency measurement point (MPH) and a value of the locus curve at the low-frequency measurement point (MPN) on the real value axis (RE) of the Nyquist diagram; and by updating a stored model parameter to the internal resistance (RG); wherein the data output interface of the parameter optimization module (11) is connected to a data input interface of the simulation module (12).

7. System (10) for optimized simulation of a characteristic of a battery cell according to claim 6, characterized by: a filter module (13) with a predictive filter algorithm for correcting a simulated state of charge (SOC) depending on the measured voltage (UM) and the simulated voltage (US).

8. System (10) for optimized simulation of a characteristic of a battery cell according to claim 6 or 7, characterized by: a measurement request module (14) for outputting a measurement request (MA) for optimization of the battery cell model (BM) depending on a condition that a change in the measured temperature (TM) exceeds a temperature change threshold, or a change in the simulated or corrected state of charge (SOC) exceeds a charge change threshold.

9. System (10) for optimized simulation of a characteristic of a battery cell according to one of claims 6 to 8, characterized by: PP34226WO / mg September 18, 2025 AVL List GmbH Means for electrochemical impedance spectroscopy measurements, comprising a pulse generator for generating measurement pulses and at least one sensor, which is associated with at least one battery cell, for measuring pulse responses.

10. Battery management system for a battery storage system with a plurality of battery cells and with a network of sensors assigned to the battery cells, characterized by: the system (10) for optimized simulation of a characteristic of a battery cell according to one of claims 6 to 9.

Images