Method and device for providing optimized charging profile of device battery by efficient parameter estimation

Abstract

The invention relates to a computer-implemented method for providing an optimized charging profile in a battery-operated engineering device (1) having a device battery (11), comprising the following steps: detecting (S2) an operating variable temporal course of an operating variable, determining (S3) operating points for parameterizing a battery model; parameterizing (S5) model parameters of the battery model by means of a fitting method and by means of the operating variable temporal progression at the determined operating point; determining (S6) a capacity-dependent aging state (SOH-C) and a resistance-dependent aging state (SOH-R) from the model parameters of the battery model; selecting (S7) a charging curve according to the capacity-dependent aging state and the resistance-dependent aging state by means of a predetermined charging curve association table, said charging curve defining the charging current of the charging process of the device battery (11) according to the charging state; and using (S8) the charging curve for at least one subsequent charging process.

Description

Technical Field

[0001] The present invention relates to providing a charging curve adapted to the aging state of a device battery. Background Art

[0002] In off-grid engineering equipment such as electric vehicles, high-performance device batteries, especially lithium-ion batteries, are used as energy storage. The electrical behavior of such device batteries is usually described by means of a battery model. The electrochemical P2D battery model is particularly suitable for lithium-ion batteries, and through the electrochemical P2D battery model, the battery behavior can be described in a particularly effective manner during use.

[0003] The P2D (pseudo-two-dimensional) battery model is a mathematical model for describing the electrochemical processes in a lithium-ion battery. The P2D battery model is capable of: predicting the performance and behavior of the battery under different operating conditions, and in particular, modeling the terminal voltage of the device battery.

[0004] The P2D model takes into account various physical and chemical processes in the battery, such as lithium-ion transport (transport of lithium ions between and within the anode and cathode materials), electron transport (transport of electrons through the anode, electrolyte, and cathode), electrochemical reactions (reactions that occur during battery charging and discharging), and heat generation and dissipation during the battery's operating time.

[0005] The P2D model uses partial differential equations to describe the processes and predict the performance of the battery under different operating conditions. The model can be used in computer simulations for optimizing battery design.

[0006] For example, with the help of the P2D battery model, the internal battery state of the device battery can be determined, from which the aging state of the device battery can be derived. In addition, the internal state can be used to generate a charging curve so that the charging process can be carried out within an aging-friendly operating range. An example of an aging-critical internal battery state is the anode overpotential, which must remain positive.

[0007] The parameterization of the model parameters of the P2D battery model is carried out by means of numerical optimization methods based on the time series of the operating variables of the device battery. Summary of the Invention

[0008] According to the present invention, there is provided a method for providing an optimized charging curve in an engineering device according to claim 1 and a corresponding device according to the co-pending independent claims.

[0009] Other design alternatives are described in the dependent claims.

[0010] According to a first aspect, a computer-implemented method for providing an optimized charging curve in a battery-operated engineering device having a device battery is proposed, having the following steps:

[0011] - Detecting the time profile of the operating variables of the operating variables;

[0012] - Determining an operating point for parameterizing the battery model;

[0013] - Parameterizing the model parameters of the battery model by means of a fitting method;

[0014] - Determining a capacity-related aging state and a resistance-related aging state from the model parameters of the battery model;

[0015] - Selecting a charging curve according to the capacity-related aging state and the resistance-related aging state by means of a preset charging curve association table, the charging curve defining a charging current for the charging process of the device battery according to the state of charge;

[0016] - Using the charging curve for at least one subsequent charging process.

[0017] In the device battery, the aging state (SOH: State of Health) is a key parameter for describing the remaining battery capacity or the remaining battery charge as the battery state of the device battery. The aging state of the device battery is usually not measured directly. This would require a series of sensors inside the device battery, which would make the production cost of such a device battery high and costly and would increase the space requirement.

[0018] The aging state is a measure of the aging of the device battery. In the case of a device battery or a battery module or a battery cell, the aging state can be described as the capacity retention rate (SOH-C). The capacity retention rate SOH-C, i.e., the capacity-related aging state, is described as the ratio of the measured instantaneous capacity to the initial capacity of the fully charged battery and decreases with increasing aging. Alternatively, the aging state can be described as the increase in the internal resistance relative to the internal resistance at the start of the service life of the device battery (SOH-R). The relative change SOH-R of the internal resistance increases with increasing battery aging.

[0019] Therefore, the aging state of the device battery is effective with respect to the battery state of the device battery, and the battery state indicates the future usage possibility of the device battery. For example, if an anomaly or fault in the device battery is to be identified early, it is also useful to know other battery states that describe the internal electrochemical cell state of the device battery. Such internal electrochemical cell states can include, for example, layer thickness (e.g., SEI thickness), description of the change in recyclable lithium due to anode / cathode side reactions, description of rapid electrolyte consumption, description of slow electrolyte consumption, description of the loss of active material in the anode, description of the loss of active material in the cathode, anode potential, etc. The aging state and internal electrochemical state of the device battery can be determined as battery states by means of an electrochemical cell model, by evaluating non-linear differential equations according to the trend of operating variables.

[0020] In addition, the battery model can correspond to a P2D battery model, where the fitting method uses the least squares method or the maximum likelihood method.

[0021] The electrochemical cell model includes a system of differential equations, which are based on differential equations parameterized by model parameters, model the internal battery state (especially the equilibrium state and, if necessary, the kinetic state) by means of a time integration method, and provide the relationship between the operating variables, battery current, battery voltage, battery temperature of the battery cells of the device battery and the state of charge of the device battery. Such an electrochemical cell model is known, for example, from the references US2016 / 023,566, US2016 / 023,567 and US2020 / 150,185.

[0022] The aging state of the device battery (not only in the form of the capacity-related aging state SOH-C but also in the form of the resistance change-related aging state SOH-R) can be approximately determined, for example, as a linear combination of the internal electrochemical cell states.

[0023] The degradation of the device battery is determined especially by the charging power during the charging process. For gentle charging, the charging process is controlled by means of a charging curve, where the maximum charging current is described according to the state of charge or the charging time. Here, the charging current usually decreases as the state of charge increases.

[0024] The charging curve of the battery is determined based on the internal battery state, and the internal battery state can be obtained with the aid of a P2D battery model. In particular, the charging curve is determined such that the anode overpotential always remains positive. The charging curves related to the aging state can be provided in the form of a charging matrix respectively, and can be implemented in the control device of the engineering equipment in which the battery operates in the form of a charging curve association table according to the aging state, such that according to the specific capacity-related and resistance change-related aging states SOH-C, SOH-R, the charging curve optimized for the current internal battery state can always be selected and used. This reduces the aging of the device battery during the charging process when the charging duration is as low as possible, and thus reduces the limitation on its usability.

[0025] The aim of the above method is: even with limited computing power, to adapt the charging curve in the control device of the engineering equipment according to the specific capacity-related and resistance change-related aging states SOH-C, SOH-R, such that the optimal charging curve can be applied. For this purpose, the P2D battery model is first parameterized in the state at the start of the life (i.e., the brand-new state of the device battery). This is usually carried out by measuring the relevant device battery on a test bench, and thus the model parameters of the battery model are determined for the initial battery state according to the battery variable trends, such as the trends of the battery current and the battery temperature, by means of a suitable parameterization method, such as a fitting method based on the RMS method (least squares method) or the maximum likelihood method.

[0026] Based on the model parameters of the parameterized battery model, the parameter space of the model parameters of the battery model can then be estimated until the end of the service life duration of the device battery. The parameter space describes the value range of the model parameters within which the relevant model parameters are allowed to move. The parameter space of the model corresponds to the physically allowed states of the electrochemical model, such as the value range of the recyclable lithium. Then the capacity-related and resistance change-related aging states SOH-C, SOH-R and the optimal charging curve are associated with the parameter combinations of the parameter space.

[0027] Then, the battery model is parameterized respectively for the varying model parameters within the parameter space. By evaluating the battery model, on the one hand, the capacity-related and resistance change-related aging states SOH-C, SOH-R are determined, and the charging curve is obtained according to the internal battery state determined via the relevant battery model. A charging curve association table is provided, which associates the capacity-related and resistance change-related aging states SOH-C, SOH-R with the respectively obtained charging curves for different parameterized battery models.

[0028] The charging curve correlation table is determined offline before the device battery is put into operation and stored in the control unit of the engineering device. Based on the parameter space determined from the model parameters, the charging curve correlation table can thus be implemented in a suitable manner in the control unit of the engineering device. Then, a suitable charging curve is selected according to the capacity-related and resistance-change-related aging states SOH-C and SOH-R determined by the control unit. In this way, the charging curve can be continuously adapted without the need for a communication connection to a central unit remote from the device. Thus, the implementation of the charging curve correlation table is a cost-effective possibility for gently operating the device battery.

[0029] The capacity-related and resistance-change-related aging states SOH-C and SOH-R can be determined by performing a fitting method on the battery model in the control unit.

[0030] It can be proposed that the charging curve is selected according to the capacity-related aging state, especially the capacity-related aging state of the most strongly aged battery cell, and according to the resistance-related aging state, especially the resistance-related aging state of the most strongly aged battery cell.

[0031] During the operation of the engineering device, the model parameters of the battery model are determined based on characteristic operating points or operating time periods, and the capacity-related aging state and resistance-related aging state are determined therefrom, and the charging curve is selected accordingly. In particular, the operating point can be determined cell-by-cell such that the capacity-related aging state SOH-C of the most aged battery cell can be determined with the aid of the balance model.

[0032] The balance model (equilibrium model) describes the part of the electrochemical model that can be estimated by the rest-phase voltage of the battery. The capacity throughput and rest voltage between the respective points can directly estimate the capacity-related aging state SOHC and the remaining capacity through a fitting process and comparison with the OCV-SOC curve at the beginning of life. This is achieved by estimating at least three balance parameters (possibly more depending on the cell chemistry). The balance parameters (equilibrium parameters) are anode volume fraction, cathode volume fraction, and cyclable lithium.

[0033] The operating points for evaluating the balance model can in particular be operating variables after the relaxation duration after the discharge phase or the charging phase.

[0034] Furthermore, an operating point for determining kinetic parameters, such as a current pulse with defined boundary conditions, can be determined. The boundary conditions are, for example, a specific battery temperature, a specific state of charge (range), a specific pulse duration, etc. Thereby, an aging state related to the resistance change can be calculated. Here, the aging state related to the resistance can also be calculated for the most aged battery cell, and the battery current of the battery cell involved can be monitored accordingly.

[0035] Determining the aging state related to the capacity by means of an equilibrium model only requires a small amount of computing power and can thus be carried out in the control unit of the engineering device. The so-called equilibrium parameters are used to determine the aging state related to the capacity of the battery cell.

[0036] The capacity throughput and the resting voltage between respective points can directly estimate the aging state SOHC related to the capacity and the remaining capacity by a fitting process and comparison with the OCV-SOC curve at the beginning of the life. This is achieved by estimating at least three equilibrium parameters. The equilibrium parameters thus determine the shape of the OCV curve, and the equilibrium parameters change strongly corresponding to the aging state SOHC related to the capacity as the cell ages. The equilibrium parameters can be estimated by a fitting process of the resting voltage points and the current throughput between its associated state of charge or points.

[0037] The kinetic parameters are determined during dynamic operation (driving or charging). The kinematic parameters are mainly associated with the aging state SoHR related to the resistance change. The calculation of the kinetic parameters is usually computationally intensive and data-intensive. Therefore, these are avoided in the method.

[0038] For a wide range of relevant aging state ranges, the equilibrium parameters have a dominant influence on the charging curve, while the change in the kinetic parameters that affect the aging state related to the resistance change only has a greater influence on the charging curve optimized therefor in the case of a smaller aging state, for example, in the case of about 85% SOH-C.

[0039] According to another aspect, a computer-implemented method for providing a charging curve association table for an individual device battery by means of the above method is proposed. The computer-implemented method has the following steps:

[0040] - At the start of the commissioning of the device battery, detect the time trend of the operating variables of the operating variables.

[0041] - Determine the model parameters of the battery model describing the device battery, wherein the aging state related to the capacity and the aging state related to the resistance can be determined from the value combinations of the model parameters.

[0042] - Provide a parameter space of the model parameters of the battery model, the parameter space specifying the value range of the model parameters.

[0043] - Create a charging curve association table that associates multiple different charging curves with combinations consisting of a capacity-related aging state and a resistance-related aging state; wherein the charging curves, the associated capacity-related aging states, and the resistance-related aging states are respectively derived from the determined value combinations of the model parameters.

[0044] According to another aspect, there is provided a device for providing an optimized charging curve in a construction device operating with a device battery, the device being designed for:

[0045] - Detect the time trend of the operating variable of the operating variable,

[0046] - Determine the operating point for parameterizing the battery model;

[0047] - Parameterize the model parameters of the battery model by means of a fitting method;

[0048] - Derive the capacity-related aging state and the resistance-related aging state from the model parameters of the battery model;

[0049] - Select a charging curve according to the capacity-related aging state and the resistance-related aging state by means of a preset charging curve association table, the charging curve defining the charging current for the charging process of the device battery according to the charging state;

[0050] - Use the charging curve for at least one subsequent charging process. Description of the Drawings

[0051] Hereinafter, the embodiments will be explained in more detail with reference to the accompanying drawings. Among them:

[0052] Figure 1 A schematic diagram showing a vehicle with a battery operation in which a charging curve is implemented;

[0053] Figure 2 A view showing an exemplary charging curve;

[0054] Figure 3 A method for operating a vehicle and a vehicle battery is shown; and

[0055] Figure 4 A method for determining a charging curve association table for use in the control unit of a vehicle is shown. Detailed Description of the Embodiments

[0056] Hereinafter, the method according to the present invention will be described based on the vehicle battery in a motor vehicle. The method for determining the charging curve is executed in the vehicle and realizes the selection of an optimized charging curve remotely generated for the vehicle according to the battery state.

[0057] Figure 1 Figure 1 shows an electrically operated vehicle 1 having a vehicle battery 11 as a rechargeable electrical energy store, an electric drive motor 12, and a control unit 13. The control unit 13 can be connected to a communication module 14 which is suitable for receiving data from a central unit 2 (so-called cloud).

[0058] A charging curve correlation table can be implemented in the control unit 13 which provides a charging curve for the charging process based on the aging state of the vehicle battery 11.

[0059] The charging curve of the vehicle battery 11 is based on the state of charge SOC [%], i.e., the stored available electrical energy of the vehicle battery 11, or a preset maximum charging current I respectively during the charging time charg . The usual charging curve is stepped, as shown for example in Figure 2 . The charging curve is preset and determined such that the anode potential does not exceed a limit value in terms of the anode overpotential.

[0060] The anode overpotential can be determined from a parameterized electrochemical model. Now, an optimization is applied here to calculate the maximum charging current for each combination of the state of charge and temperature such that the anode potential at the electrode surface is greater than or equal to 0 plus an optional safety margin with respect to the potential of metallic lithium to account for model or measurement inaccuracies.

[0061] Figure 3 A method for operating the vehicle 1 is described. The method can be implemented as software and / or hardware in the control unit 13.

[0062] For this purpose, in step S1, a charging curve correlation table is provided which correlates the capacity-related aging state and the resistance change-related aging state with a specific charging curve. The charging curve can be preset as a charging profile as the correlation of the charging current with the state of charge.

[0063]

[0064] where C corresponds to a preset current value.

[0065] In step S2, the operating variables of the device battery are continuously detected at the cell level.

[0066] In a suitable operating point, especially corresponding to a rest point or a specific pulse excitation, the operating variable points and the operating variable trends at the cell level can be stored in step S3. The rest point corresponds to the operating variables, especially after a relaxation phase between, for example, 15 and 30 minutes after the charging and discharging phases, and is used to determine the equilibrium parameters of an equilibrium model which is part of the battery model or the entire battery model.

[0067] The equilibrium model describes the part of the electrochemical model that can be estimated by the open-circuit voltage during the rest phase of the battery. Via the capacity throughput and the open-circuit voltage between individual points, the capacity-related state of health SOHC and the remaining capacity can be directly estimated through a fitting process and comparison with the OCV-SOC curve at the start of life. This is achieved by estimating at least three equilibrium parameters (possibly more depending on the cell chemistry). The equilibrium parameters are the anode volume fraction, the cathode volume fraction, and the cyclable lithium.

[0068] Furthermore, in step S4, the trend of the operating variables can be monitored for a current pulse with defined boundary conditions, which are determined, for example, by a specific battery temperature, a specific state-of-charge range (e.g., between 40 - 60% SOC), and a specific d-pulse duration of, for example, 1 - 5 seconds. The determination of the operating point and the subsequent evaluation are carried out during a specific time period before the current time point. The storage of the rest point and the current pulse operation can be performed in the control unit 13.

[0069] Then, in step S5, the operating point can be evaluated with the aid of a suitable battery model in order to determine the model parameters of the battery model.

[0070] The P2D (pseudo-two-dimensional) battery model is a mathematical model for describing the electrochemical processes of a lithium-ion battery. The P2D battery model is capable of: predicting the performance and behavior of the battery under different operating conditions, and in particular, modeling the terminal voltage of the device battery.

[0071] Common P2D models take into account various physical and chemical processes in the battery, such as lithium-ion transport (transport of lithium ions between and within the anode and cathode materials), electron transport (transport of electrons through the anode, electrolyte, and cathode), electrochemical reactions (reactions occurring during battery charging and discharging), and heat generation and dissipation during battery operation.

[0072] The P2D model uses partial differential equations to describe the processes. The parameterization of the model parameters of the P2D battery model is based on the operating point by means of a fitting method based on the least squares method or the maximum likelihood method.

[0073] For example, with the aid of the P2D battery model, the internal state of the device battery can be determined, from which the aging state of the device battery can be derived.

[0074] In step S6, the aging states SOH-R related to the current resistance change and SOH-C related to the capacity of each battery cell are determined from the model parameters. This can be done, for example (but not exclusively), by comparing the resting voltage with the current integration of the capacity-related aging state SoHC and the voltage response to a current pulse of the resistance change-related aging state.

[0075] Next, in step S7, with the aid of the provided charging curve correlation table, a suitable charging curve is selected based on the capacity-related aging state of the most aged battery cell and based on the resistance change-related aging state of the most aged battery cell.

[0076] In step S8, this charging curve is used for subsequent charging processes. The method adapts cyclically and continuously to the current battery state.

[0077] The determination of the charging curve correlation table can be obtained separately before the vehicle battery is put into operation. According to Figure 4 the flowchart in

[0078] In step S11, for this purpose, the battery model is first parameterized for the individual device battery 11. This can be done by measuring the trend of the operating variables on a test bench and then parameterizing the battery model.

[0079] In step S12, the possible parameter space of the model parameters can be estimated until the end of the service life of the vehicle battery. This applies both to the balancing parameters, which significantly affect the capacity-related aging state, and to the dynamic (kinetic) model parameters of the battery model, which have a significant impact on the resistance-related aging state of the battery cell. The parameter space can be estimated by empirical values of similar cell chemistries. The empirical values can be taken from the literature or from laboratory or field tests carried out by oneself.

[0080] Next, in step S13, now, with the aid of the electrochemical battery model, the smoothest charging curve is obtained for various combinations of the capacity-related aging state and the resistance change-related aging state, and the charging curve is associated with the corresponding combination. Since a large number of electrochemical states and thus various charging curves can be associated with each combination of the capacity-related aging state and the resistance change-related aging states SoHC and SoHR, as a conservative approach, the lowest current for each charge state-temperature combination is selected separately from all the charging curves.

[0081] Now, in the subsequent step S14, the charging curve can be implemented in the control unit of the vehicle.

Claims

1. A computer-implemented method for providing an optimized charging curve in a battery-operated engineering device (1) having a device battery (11), the method comprising the following steps: - detect (S2) the time trend of the operating variables of the operating variables, - determining (S3) an operating point for parameterizing the battery model; - parameterizing the model parameters of the battery model by means of a fitting method using the time profile of the operating variable at the determined operating point (S5); - determining (S6) a capacity-related aging state (SOH-C) and a resistance-related aging state (SOH-R) from the model parameters of the battery model; - selecting (S7) a charging curve according to the capacity-related aging state and the resistance-related aging state by means of a preset charging curve association table, the charging curve defining a charging current for a charging process of the device battery (11) according to the charging state; as well as - Using the charging profile (S8) for at least one subsequent charging process. 2 . The method according to claim 1 , wherein the battery model corresponds to a P2D battery model, wherein the fitting method uses a least squares method or a maximum likelihood method.

3. A method according to claim 1 or 2, wherein the charging curve is selected according to the capacity-related aging state (SOH-C), in particular the capacity-related aging state with respect to the most aged battery cell, and according to the resistance-related aging state (SOH-R), in particular the resistance-related aging state with respect to the most aged battery cell.

4. A computer-implemented method for providing a charging curve association table for a single device battery (11) by means of a method according to any one of claims 1 to 3, the computer-implemented method comprising the following steps: - at the start of operation of the device battery (11), detecting (S11) the temporal trend of the operating variable of the operating variable, - determining (S3) model parameters of a battery model describing the device battery (11), wherein a capacity-related aging state (SOH-C) and a resistance-related aging state (SOH-R) can be determined from a combination of values ​​of the model parameters; - providing a parameter space of a model parameter of the battery model, the parameter space indicating a value range of the model parameter; - Creating (S13) the charging curve association table, which associates a plurality of different charging curves with a combination of a capacity-related aging state and a resistance-related aging state, respectively; wherein the charging curve, the associated capacity-related aging state and the resistance-related aging state are respectively derived from the determined value combination of the model parameters.

5. A device for providing an optimized charging curve in a battery-operated engineering device (1) having a device battery (11), wherein the device is designed to: - Detect the time trend of running variables, - Determine the operating point for parameterizing the battery model; - parameterizing the model parameters of the battery model by means of a fitting method; - determining a capacity-related aging state and a resistance-related aging state from the model parameters of the battery model; - selecting a charging curve according to the capacity-related aging state and the resistance-related aging state by means of a preset charging curve association table, the charging curve defining a charging current for a charging process of the device battery according to the charging state; - using the charging profile for at least one subsequent charging process.

6. A computer program product comprising instructions which, when the program is executed by at least one data processing device, cause the data processing device to execute the steps of the method according to any one of claims 1 to 5.

7. A machine-readable storage medium comprising instructions, which, when executed by at least one data processing device, cause the data processing device to perform the steps of the method according to any one of claims 1 to 5.

Images