Method and apparatus for determining the aging state of a device battery based on a voltage curve during a relaxation phase

Abstract

Method, in particular a method, at least partially computer-implemented, for determining the state of aging (SOH-C) of a device battery (2) in a technical device (1), comprising the following steps: - Recording (S5) a voltage waveform of the battery voltage of the device battery (2) during a relaxation phase; - Selecting a reference voltage profile depending on a battery state at the beginning of the relaxation phase (RP); - Determining (S8) an integral value of a differential voltage waveform between the reference voltage waveform and the recorded voltage waveform; - Determining (S9) an aging state depending on the integral value using an assignment model that is assigned to the reference voltage curve and is designed to assign an aging state of the device battery (2) to each integral value.

Description

Technical field

[0001] The invention relates to electrical devices operated independently of the mains with device batteries, in particular electrically powered motor vehicles, especially electric vehicles or hybrid vehicles, and furthermore to measures for determining a current aging state (SoH: State of Health) of the device battery. Technical background

[0002] The energy supply for off-grid electrical devices and machines, such as electric vehicles, is provided by device batteries or vehicle batteries. These supply electrical energy to operate the devices or vehicles.

[0003] Device batteries degrade over their lifespan, depending on their load and usage. This so-called aging leads to a continuously decreasing maximum power or storage capacity. The state of aging is a measure used to indicate the aging of device batteries. By convention, a new device battery has a state of aging of 100% of its available total capacity, which decreases progressively over its lifespan. A measure of device battery aging (the change in the state of aging over time) depends on the individual load on the device battery; for vehicle batteries, this includes the driver's usage patterns, external environmental conditions, and the vehicle battery type.

[0004] The state of health (SOH) of a device battery, the value of which is calculated and stored in a battery control unit (BCU), is read out via the OBD interface using a diagnostic tool during inspection in a car repair shop in the case of a vehicle battery. These state of health values ​​were previously determined by the battery control unit during or after driving, during or after charging, in dynamic or stationary states, under different conditions, and depending on the algorithm used to determine the state of health.

[0005] A standardized method for determining the state of aging is currently unknown. Generally, the state of aging read from the battery control unit is relied upon, but this has a high degree of inaccuracy (~5%). A more precise value for the state of aging offers significant advantages and planning certainty for estimating the residual value of the device battery.

[0006] Connecting to a central processing unit (CPU), such as a cloud, combined with the execution of data-intensive algorithms, enables a more precise and robust determination of aging states. However, this method requires a communication link with the CPU, which is not always available.

[0007] Document DE 10 2021 113 456 A1 discloses a method and a device for non-invasively determining a battery, with which the battery can be characterized using parameters based on the analysis of relaxation voltages after a charging process.

[0008] Document DE 10 2022 200 345 A1 discloses a battery with an integrated condition monitoring function at the level of individual battery cells or small cell groups. A special hardware architecture at the cell level enables isolated measurement of individual cells during operation.

[0009] Document DE 10 2022 120 011 A1 discloses a method for estimating the state of charge (SoC) of battery cells within a reconfigurable battery during operation, without having to wait for the lengthy relaxation time.

[0010] The object of the present invention is to provide an alternative method for determining an aging state that can be carried out in a simple manner and with low data and computational intensity. Disclosure of the invention

[0011] This problem is solved by the method for determining the aging state of a device battery of a technical device according to claim 1, as well as a corresponding device and a battery system according to the dependent claims.

[0012] Further details are specified in the dependent claims.

[0013] According to a first aspect, a computer-implemented method for determining the aging state of a device battery in a technical device is provided, comprising the following steps: - Recording a voltage curve of the device battery during a relaxation phase; - Selecting a reference voltage profile depending on the battery state at the beginning of the relaxation phase; - Determining an integral value of a differential voltage curve between the reference voltage curve and the recorded voltage curve; - Determining an aging state depending on the integral value using an assignment model that is assigned to the reference voltage curve and is designed to assign an aging state of the device battery to each integral value.

[0014] The aging state of a device battery is not usually measured directly. This would require opening the battery cells and measuring them in a test bench, or alternatively, installing a series of sensors inside the device battery. This would make manufacturing such a device battery expensive and complex, and would increase the space required. Furthermore, practical measurement methods for directly determining the aging state of device batteries are not yet available on the market.

[0015] The battery's state of aging is currently determined in a battery control unit located near the battery and read out during inspection or maintenance. The provided state of aging data is determined under varying conditions during driving or after charging, depending on the algorithm used for determining the state of aging. The methods used in the battery control unit of the battery management system can vary considerably, making it generally impossible to compare the state of aging data read from different battery control units across vehicles. Furthermore, inaccuracies can be as high as 5%. Accurate determination of the state of aging is important for the device user, as it can determine the remaining battery life and, consequently, the future usability of the device.

[0016] The state of health (SOH) is a key parameter for determining the remaining total battery capacity or remaining range at full charge in device batteries. The state of health represents a measure of the battery's aging. In the case of a device battery, battery module, or battery cell, the state of health can be expressed as the capacity retention rate (SOH-C). The SOH-C is defined as the ratio of the measured instantaneous capacity to the initial capacity of the fully charged battery. This ratio decreases with age. Alternatively, the state of health can be expressed as the increase in internal resistance (SOH-R) relative to the initial resistance of the device battery. The ratio of current resistance to initial resistance, and thus the SOH-R, increases as the battery ages.

[0017] One way to determine the aging state of a device battery is to use the relationship between the battery's open-circuit voltage (i.e., the open-circuit voltage) after a sufficiently long relaxation period (the time after the last current flow into or out of the device battery) and the device battery's state of charge. The state of charge indicates the electrical charge stored and available in the device battery relative to its total storable charge. The relationship between the battery's open-circuit voltages after each sufficiently long relaxation period (up to a point where the time gradient of the battery voltage is less than a predetermined threshold, e.g., up to a point where there is no change in voltage of 0.5 mV / 10 s for a defined period of, for example, 10 seconds) allows for the determination of the battery's state of charge.(occurred after 30 minutes), and the corresponding state of charge corresponds to an open-circuit voltage characteristic or OCV characteristic (OCV: Open Circuit Voltage), which is provided as an OCV model. While the OCV models known to date describe the dependence of the open-circuit voltage on the state of charge and, if applicable, the temperature of the battery, they do not consider the dependence on the battery's aging state or the path dependence of the aging effects.

[0018] Inaccurate determination of a device battery's aging state leads to an inaccurate calculation of its remaining service life, which can be determined by predicting the expected aging process. This can result in the device battery being replaced prematurely or being operated for too long with insufficient performance. The method described above involves evaluating the voltage profile of the battery (terminal voltage) during a relaxation phase to determine the device battery's aging state. This approach is based on the fact that the voltage profile during the relaxation phase of a device battery is significantly dependent on its aging state.

[0019] It can be provided that the start of the relaxation phase is detected when the battery current is 0 A, and the end of the relaxation phase is detected when the gradient of the battery voltage falls below a predetermined threshold.

[0020] The relaxation phase of a device battery generally corresponds to the period during which the battery voltage changes after being subjected to a charging or discharging current. The beginning of the relaxation phase can be defined by the start of a period with a current flow of 0 A, and the end of the relaxation phase by the point in time at which the magnitude of the change in battery voltage falls below a predetermined threshold, such as 0.5 mV / 10 s.

[0021] It may be stipulated that the procedure is only executed if a discharge phase or a charging process has taken place immediately before the relaxation phase, which has led to a change in the state of charge of more than a predetermined threshold amount.

[0022] To determine the state of aging, the recorded battery voltage profile within the relaxation phase of the device battery under test is first determined. A time-based reference voltage profile is then provided, indicating the voltage profile of the battery (without current) during a relaxation phase for a non-aged device battery. The area between the measured voltage profile and the reference voltage profile during the relaxation phase is calculated. Using a simple mapping table or mapping function, this calculated area can then be assigned to the state of aging of the device battery.

[0023] The determination of the battery's aging state is based on the relationship between the area of ​​the voltage difference resulting from the relaxation voltage profile of the device battery and a reference voltage profile specific to the battery type. This method has the advantage of being independent of the battery's usage pattern, as it is performed during the battery's relaxation phase, i.e., without any load on the battery, thus ensuring reliable and reproducible conditions. Furthermore, the method requires minimal data, as no historical operating parameters need to be considered.

[0024] It may be provided that the battery status is determined by at least one of the following parameters: - a current charge level at the beginning of the relaxation phase, - a current battery temperature at the beginning of the relaxation phase, - an energy throughput during a predetermined period immediately preceding a charging or discharging phase, - the current battery voltage at the beginning of the relaxation phase; and - information about the cell chemistry of the device battery.

[0025] The reference voltage curve for an unaged device battery can be determined based on its state of charge. The battery state can be defined by specifying the cell chemistry and at least one of the following parameters: battery temperature, final charging voltage, and average charging or discharging power within a given period before the start of the relaxation phase. This allows the area between the measured relaxation voltage curve of a device battery under test and the reference voltage curve to be determined according to the battery state of charge.

[0026] The above method makes it possible to determine the aging state based on the relaxation phase of a device battery, without having to specify concrete times for the start or end of the measurement. To determine the area, the relevant reference voltage curve is selected depending on the battery's state of charge, and the area is calculated. Using the assignment model, an aging state can then be assigned to this area.

[0027] The assignment model for an aged device battery whose state of aging is known can be determined by assigning the area between the measured voltage waveform and the reference voltage waveform associated with the battery state to the state of aging. This requires determining the assignment model separately for each of the reference voltage waveforms. The states of aging determined by the assignment model can be obtained, for example, using other methods for determining the state of aging. Other methods for determining the state of aging might include, for example, a Coulomb counting method performed during battery maintenance or a time-series integration-based state-of-aging method performed in a remote central processing unit.

[0028] It may be provided that the procedure is carried out internally by the device, with the reference voltage waveforms and the associated mapping model being transmitted from a remote central unit to the technical device.

[0029] Furthermore, the aging state determined in this way can be signaled or used to determine a remaining service life, in particular by extrapolating a temporal progression of the aging state in order to determine the end of the service life as the point in time at which a predetermined threshold value of the aging state is reached.

[0030] According to another aspect, a procedure for creating an allocation model is provided, with the following steps: - Recording voltage curves for a new, unused device battery for different battery states during relaxation phases; - Storing the voltage curves as reference voltage curves assigned to the respective battery state; - For multiple aged device batteries, recording battery states at the beginning of relaxation phases and voltage profiles during the relaxation phase in question; - Determining an aging state using a method for determining an aging state essentially simultaneously with the recording of the respective stress profile; - Determining an integral value of a differential voltage curve between the recorded voltage curve for the aged device battery in question and the reference voltage curve associated with the recorded battery state; and - Determining support points for the assignment model for the reference voltage curve assigned to the recorded battery state, where the support points each specify an assignment of the integral value to the aging state; - Creating the allocation model using the support points.

[0031] According to another aspect, a device for carrying out the above procedure is provided. Brief description of the drawings

[0032] The embodiments are explained in more detail below with reference to the accompanying drawings. These show: Fig. 1 a schematic representation of a vehicle with a vehicle battery and a battery control unit in which the procedure for determining the state of aging is carried out; Fig. 2. A flowchart illustrating the steps of the procedure for determining the state of aging in the vehicle's battery control unit; Fig. 3. A flowchart to illustrate the steps for creating an allocation model; Fig. 4 Voltage curves during a relaxation phase for batteries of different ages after a charging process; and Fig. 5. A voltage curve during a relaxation phase after a discharge process. Description of embodiments

[0033] The inventive method is described below using a vehicle battery as an example of a device battery in a motor vehicle. The vehicle battery can be measured during a relaxation phase after charging at a charging station (as an example of a power supply station) to determine its state of aging. Measuring during the relaxation phase allows for the maintenance of reproducible conditions, such as a constant battery temperature, so that the state of aging can be determined with particular accuracy using a suitable method for determining the state of aging.

[0034] The above example is representative of a wide variety of stationary or mobile devices with off-grid power supply, such as vehicles (electric vehicles, pedelecs, etc.), systems, machine tools, household appliances, IoT devices, and the like.

[0035] Fig. Figure 1 shows a vehicle 1, in particular an electric vehicle, with a vehicle battery 2 for operating a drive motor 3. The operation of the drive motor 3 is controlled by a control unit 5; that is, the power drawn from the vehicle battery 2 or the energy fed back into the vehicle battery 2 due to recuperation operations is controlled according to driver inputs or inputs from other vehicle systems. The vehicle can be connected to a charging station 3 via a charging port 6 to supply electrical energy for charging the vehicle battery 2.

[0036] A battery control unit 4 is provided, which controls the functions of the vehicle battery 2 and records measurement data during the operation of the vehicle battery 2. Thus, during operation of the vehicle 1 or during the charging process, the energy consumption and / or energy input of the vehicle battery 2 is continuously recorded in order to capture and temporarily store a time-dependent profile of the battery voltage, particularly at the pack, module, and / or cell level. Furthermore, the amount of energy consumed or supplied can be temporarily stored in the form of a time-dependent profile of the battery current. The temperature of the vehicle battery 2 can also be measured and its profile recorded.

[0037] To monitor battery aging and determine its remaining lifespan, the current aging status can be determined regularly. The battery control unit executes the procedure as described in the flowchart of the Fig. 2 will be explained in more detail.

[0038] In step S1, it is checked whether a battery current of 0 A is flowing. If this is the case (alternative: Yes), the procedure continues with step S2; otherwise (alternative: No), it jumps back to step S1.

[0039] In step S2, it is checked whether a previous charging or discharging phase of a continuous current flow has led to a change in the state of charge exceeding a predefined threshold, in particular a change of more than 20% of the state of charge. This change could have occurred due to driving or charging. If this is the case (alternative: Yes), the procedure continues with step S3; otherwise (alternative: No), the procedure is terminated or restarted by returning to step S1.

[0040] In step S3, parameters that characterize the battery status of the vehicle battery are recorded, such as the current state of charge, the current battery temperature, the power throughput during a predetermined period immediately preceding a charging or discharging phase, the current battery voltage, and the like.

[0041] In step S4, a reference voltage curve is identified and selected depending on the battery's state of charge. For this purpose, the battery control unit stores reference voltage curves for the relaxation phases of a new vehicle battery, i.e., a vehicle battery in its state of charge immediately after manufacturing. From these reference voltage curves, a specific curve is selected based on the battery's state of charge.

[0042] In a subsequent step S5, a battery voltage is recorded at time intervals (corresponding to a predefined sampling rate) of between 0.1 and 5s and the course of the battery voltage is stored.

[0043] In step S6, it is checked whether the battery current is still 0 A. If this is the case (alternative: Yes), the procedure continues with step S7; otherwise (alternative: No), the procedure is aborted and the process returns to step S1.

[0044] In step S7, it is checked whether the time gradient of the measured battery voltage falls below a predetermined threshold, such as 0.5 mV / 10 s. If this is the case (alternative: yes), the procedure continues with step S8; otherwise, it returns to step S4. The battery voltage profile is determined in this way for the duration of the relaxation phase.

[0045] Subsequently, in step S8, a differential voltage profile between the reference voltage profile and the recorded battery voltage profile is determined and integrated over the duration of the relaxation phase to obtain an area (unit of area size V*s) with respect to the profile of the differential voltage.

[0046] The magnitude of the integrated differential stress is evaluated in step S9 using a mapping model. This mapping model is assigned to the previously selected reference stress profile and is designed to assign the determined area of ​​the differential stress profile to an aging state.

[0047] This aging state is then signaled in step S10 or used in some other way for the operation of the vehicle battery 1, for example to limit fast charging processes if the vehicle battery 1 ages too much.

[0048] The following procedure can be used to create the assignment model.

[0049] In step S11, voltage waveforms for a new, unused device battery are recorded for various battery states during relaxation phases. To create the reference voltage waveforms, particularly for an unaged, i.e., brand-new, battery, the battery voltage waveforms during relaxation phases are recorded for both charging and discharging processes preceding the relaxation phase. Parameters preferably used for the battery state include the battery temperature at the beginning of the relaxation phase, the energy throughput of the preceding charging or discharging phase within a predetermined period before the start of the relaxation phase, the current state of charge at the beginning of the relaxation phase, the change in the state of charge during the preceding charging or discharging phase, the relaxation start voltage, and similar parameters.Thus, a number of reference voltage waveforms can be recorded, which form the basis for determining the differential voltage waveform.

[0050] In step S12, these voltage curves are assigned to the respective battery state and stored as reference voltage curves.

[0051] In step S13, the battery states at the beginning of relaxation phases and voltage profiles during the respective relaxation phase are recorded for a large number of aged device batteries.

[0052] In step S14, an aging state is determined for each of the aged batteries using a method for determining an aging state, essentially simultaneously with the recording of the respective voltage profile. For each of the reference voltage profiles, an aging state can thus be determined for the creation of the assignment model for aged batteries using a different method (colomb counting or an electrochemical battery model).

[0053] Subsequently, for each aged vehicle battery, in step S15 an integral value of a differential voltage curve is determined between the recorded voltage curve for the aged device battery in question and the reference voltage curve that is assigned to the recorded battery state.

[0054] In step S16, support points for the assignment model for the reference voltage curve assigned to the recorded battery state are determined, whereby the support points each specify an assignment of the integral value to the aging state.

[0055] In step S17, the assignment model is determined using the support points. The assignment model can be a parametric regression model or a data-driven model, such as a Gaussian process model.

[0056] Further development of the allocation model can take place at any time when a precise determination of an aging state is carried out using another precise method, by conducting a relaxation phase following a charging or discharging phase with a sufficiently high change in the state of charge in close temporal proximity or immediately before or immediately after the determination of the aging state.

[0057] Fig. Figure 4 shows, for example, the curves of the reference battery voltage KR (at SOH 100%) and exemplary voltage curves measured during a relaxation phase RP after a charging process LP for different aging states SOH.

[0058] Analogously shows Fig. 5 Reference battery voltage curves KR (at SOH 100%) and exemplary measured voltage curves during a relaxation phase RP after a discharge phase EP for different aging states SOH.

Claims

[1] Method, in particular a method, at least partially computer-implemented, for determining the state of aging (SOH-C) of a device battery (2) in a technical device (1), comprising the following steps: - Recording (S5) a voltage waveform of the battery voltage of the device battery (2) during a relaxation phase; - Selecting a reference voltage profile depending on a battery state at the beginning of the relaxation phase (RP); - Determining (S8) an integral value of a differential voltage waveform between the reference voltage waveform and the recorded voltage waveform; - Determining (S9) an aging state depending on the integral value using an assignment model that is assigned to the reference voltage curve and is designed to assign an aging state of the device battery (2) to each integral value. [2] Method according to claim 1, wherein the beginning of the relaxation phase (RP) is detected when the battery current is 0 A, and wherein the end of the relaxation phase (RP) is detected when the gradient of the battery voltage falls below a predetermined threshold. [3] Method according to claim 1 or 2, wherein the method is only carried out if a discharge phase or a charging process has taken place immediately before the relaxation phase, which has led to a change in the state of charge of more than a predetermined threshold amount. [4] Method according to any one of claims 1 to 3, wherein the battery state is determined by at least one of the following parameters: - a current charge level at the beginning of the relaxation phase, - a current battery temperature at the beginning of the relaxation phase, - an energy throughput during a predetermined period immediately preceding a charging or discharging phase, - the current battery voltage at the beginning of the relaxation phase (RP); and - information on the cell chemistry of the device battery (2). [5] Method according to any one of claims 1 to 4, wherein the method is carried out internally by the device, wherein the reference voltage waveforms and the associated mapping model are transmitted from a remote central unit to the technical device (2). [6] Method according to any one of claims 1 to 5, wherein the aging state thus determined is signaled (S10) or used to determine a remaining service life, in particular by extrapolating a time course of the aging state in order to determine as the end of service the time at which a predetermined threshold value of the aging state is reached. [7] Method according to any one of claims 1 to 6, wherein the assignment model is data-based. [8] Procedure for creating an allocation model, comprising the following steps: - Recording (S11) voltage waveforms for a new, not yet operated device battery for different battery states during relaxation phases; - Storing (S12) the voltage waveforms as reference voltage waveforms assigned to the respective battery state; - For multiple aged device batteries, recording (S13) battery states at the beginning of relaxation phases and voltage profiles during the relaxation phase in question; - Determining (S14) a state of aging using a method for determining a state of aging essentially at the same time as the recording of the respective stress profile; - Determining (S15) an integral value of a differential voltage curve between the recorded voltage curve for the aged device battery in question and the reference voltage curve associated with the recorded battery state; and - Determining (S16) support points for the assignment model for the reference voltage profile assigned to the recorded battery state, wherein the support points each indicate an assignment of the integral value to the aging state; - Creating or training (S17) the assignment model using the support points. [9] Apparatus for carrying out one of the methods according to any one of claims 1 to 8. [10] Computer program product comprising instructions which, when the program is executed by at least one data processing device, cause it to perform the steps of the method according to any one of claims 1 to 8. [11] Machine-readable storage medium comprising instructions which, when executed by at least one data processing device, cause it to perform the steps of the method according to any one of claims 1 to 8.

Images