Battery management device

Abstract

The invention relates to a battery management device. This battery management device is provided with: a calculation unit that calculates SOC and a coefficient from measured values of the current and voltage of a battery on the basis of an estimation model that estimates the polarization voltage from the SOC of the battery and the coefficient relating to the amount of increase in the voltage corresponding to the SOC among the polarization voltages generated by the discharge of the battery; and a notification unit that, when the SOC is equal to or less than a threshold value, performs a notification to urge a user to charge the battery, and the notification unit decreases the threshold value as the coefficient decreases.

Description

Technical Field

[0001] The invention relates to a battery management device. Background Art

[0002] For example, Japanese Patent Application Laid-Open No. 2018-77199 discloses a technology for estimating the battery's state of charge (SOC) using an equivalent circuit model of a vehicle battery for battery management. The SOC serves as a benchmark for urging users to charge their batteries. Summary of the Invention

[0003] For example, when lithium-ion batteries are discharged, polarization causes an increase in diffusion resistance in low SOC regions (e.g., below 20%). This increase in diffusion resistance reduces the battery's available power. Therefore, even if the user is urged to charge due to a decrease in SOC, there is a possibility that the battery may have excess power, or conversely, a power shortage that could hinder load operation.

[0004] Therefore, the present invention provides a battery management device that can appropriately urge a user to charge a battery.

[0005] The battery management device of the present invention has:

[0006] a calculation unit that calculates an SOC and a coefficient from each measured value of a current and a voltage of a battery based on an estimation model, wherein the estimation model estimates the polarization voltage based on the SOC of the battery and the coefficient associated with an increase in voltage corresponding to the SOC in a polarization voltage generated by discharge of the battery; and

[0007] a notification unit that, when the SOC is below a threshold, issues a notification urging the user to charge the battery;

[0008] The notification unit lowers the threshold value as the coefficient becomes smaller.

[0009] In the above-mentioned battery management device, the notification unit may change the threshold value so as to establish a linear relationship with the coefficient.

[0010] In the battery management device described above, the calculation unit may periodically calculate the battery voltage using the estimation model and calculate the SOC and the coefficient so that a difference between a calculated value of the battery voltage and a measured value of the battery voltage converges.

[0011] In the above-mentioned battery management device, the battery may be a lithium-ion battery.

[0012] In the above-mentioned battery management device, the battery may supply electric power to an electric motor that drives the vehicle.

[0013] According to the present invention, it is possible to appropriately urge the user to charge the battery. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Features, advantages, and technical and industrial significance of exemplary embodiments of the present invention will be described below with reference to the accompanying drawings, in which like figures represent like elements, and in which:

[0015] Figure 1 FIG. 1 is a schematic diagram showing an example of a battery system of a vehicle V;

[0016] Figure 2 1 is a diagram showing an example of changes in diffusion resistance of a lithium-ion battery relative to SOC during charge and discharge;

[0017] Figure 3A This is a functional block diagram showing an example of a state calculation unit;

[0018] Figure 3B is a diagram showing an example of a temporal change in the SOC of a battery cell during discharge;

[0019] Figure 4 1 is a diagram showing an example of a change in discharge polarization voltage with respect to SOC;

[0020] Figure 5 is a diagram showing an example of a threshold map; and

[0021] Figure 6 This is a flowchart showing an example of the operation of the control device. DETAILED DESCRIPTION

[0022] Battery system structure

[0023] Figure 1 This is a schematic diagram showing an example of a battery system for a vehicle V. The vehicle V includes a control device 1 comprising one or more ECUs (Electronic Control Units) and a battery pack 2 serving as a power source. The vehicle V also includes an onboard display 3 that displays various information for passengers (users), an inlet 6 for charging and discharging the battery pack 2, an inverter 4 that converts DC current into AC current, and an electric motor (MG) 5 serving as the vehicle V's power source.

[0024] The battery pack 2 includes a battery cell 20 as an example of a battery, a current sensor 21 for detecting the current flowing through the battery cell 20, a voltage sensor 22 for detecting the voltage of the battery cell 20, and a temperature sensor 23 for detecting the temperature T of the battery cell 20. The detection values ​​of the current sensor 21, the voltage sensor 22, and the temperature sensor 23 are output to the control device 1. An example of the battery cell 20 is a lithium-ion battery, but the present invention is not limited thereto.

[0025] The control device 1 is an example of a battery management device. The control device 1 is a computer including a CPU (Central Processing Unit), ROM (Read Only Memory), and RAM (Random Access Memory). The ECU 1 operates the CPU according to a program stored in the ROM.

[0026] The control device 1 controls the inverter 4. The inverter 4 is connected between the battery pack 2 and the electric motor 5. The inverter 4 converts direct current into alternating current by switching multiple switching elements, such as MOS-FETs (Metal-Oxide-Semiconductor Field Effect Transistors). The control device 1 outputs a PWM (Pulse Width Modulation) signal with a duty cycle corresponding to the driving state of the vehicle V to the inverter 4.

[0027] The control device 1 also controls the charging of the battery unit 20. The battery unit 20 comprises, for example, a plurality of lithium-ion batteries connected in series. The battery unit 20 supplies DC power to the inverter 4, thereby driving the motor 5. The battery unit 20 is charged from an external charging station 9 via the inlet 6. The control device 1, for example, instructs the charging station 9 on a target current value to be output during charging.

[0028] Control device 1 includes an operation control unit 10, a capacity estimation unit 11, a state calculation unit 12, and a charge notification unit 13 as software functions implemented by program execution. It should be noted that operation control unit 10, capacity estimation unit 11, state calculation unit 12, and charge notification unit 13 may also be implemented as hardware such as an integrated circuit.

[0029] The control device 1 also includes a nonvolatile memory 14 such as an EEPROM (Electrically Erasable Programmable ROM). The memory 14 stores OCV (Open Circuit Voltage) map data (OCV map) 140 and threshold map data (threshold map) 141.

[0030] The OCV map 140 and the threshold value map 141 are generated based on experimental results and simulation results of the characteristics of the battery cell 20 and are pre-written in the memory 14. The OCV map 140 shows the correlation between the OCV and the SOC of the battery cell 20. The threshold value map 141 shows the correlation between the SOC threshold for urging charging of the battery cell 20 and the discharge polarization coefficient described later.

[0031] The operation control unit 10 instructs the capacity estimating unit 11, the state calculating unit 12, and the charge notifying unit 13 to operate in a predetermined sequence. The operation control unit 10 also controls the charging station 9 and the inverter 4 during charging and discharging.

[0032] Cap=100÷(SOCe-SOCs)×Is…(1)

[0033] The capacity estimation unit 11 estimates the full charge capacity Cap of the battery cell 20, for example, during charging at the charging station 9. As an example, the capacity estimation unit 11 calculates the full charge capacity Cap according to the above-mentioned formula (1). In the formula (1), Is is the cumulative value after the current value of the battery cell 20 is accumulated for a predetermined period. In the formula (1), SOCs is the SOC of the battery cell 20 when the accumulation of the current of the battery cell 20 is started, and SOCe is the SOC of the battery cell 20 when the accumulation of the current is ended. The capacity estimation unit 11 calculates the cumulative value Is based on the detection value of the current sensor, for example, and obtains the detection value of the voltage sensor 22 during charging, that is, SOCe and SOCs corresponding to the OCV, from the OCV map 140.

[0034] The state calculation unit 12 is an example of a calculation unit. The state calculation unit 12 calculates the SOC and discharge polarization coefficient from the measured current and voltage values ​​of the battery cell 20 based on an estimation model. The estimation model estimates the discharge polarization voltage based on the SOC of the battery cell 20 and the discharge polarization coefficient, which is correlated with the increase in voltage corresponding to the SOC in the polarization voltage (discharge polarization voltage) generated by the discharge of the battery cell 20. The measured current and voltage values ​​are obtained, for example, as detected by the current sensor 21 and the voltage sensor 22. As described later, the estimation model is constructed based on, for example, a circuit equation for the closed circuit voltage (CCV) of the battery cell 20.

[0035] The charging notification unit 13 is an example of a notification unit. When the SOC of the battery unit 20 falls below a threshold, the charging notification unit 13 notifies the user of the vehicle V to urge charging of the battery unit 20. The charging notification unit 13 obtains the SOC and discharge polarization coefficient calculated by the state calculation unit 12 via the operation control unit 10 and obtains a threshold value corresponding to the discharge polarization coefficient from the threshold map 141. The charging notification unit 13 uses the threshold value corresponding to the discharge polarization coefficient to determine whether charging is necessary. As described later, this allows the user to be appropriately urged to charge according to the usage status of the battery unit 20.

[0036] When the SOC is below a threshold, the charging notification unit 13 displays a charging notification, such as characters or symbols, on the onboard display 3. This allows the user to recognize the need to charge the battery unit 20 at the appropriate time and charge the battery unit 20 before it reaches a state of insufficient power and becomes insufficient. It should be noted that while the onboard display 3 is used as a means for notifying the user of charging in this example, the present invention is not limited to this. Other means, such as audio notification using a speaker or light emission from an LED (Light Emitting Diode), may also be used.

[0037] Processing of the state calculation unit

[0038]

[0039] The state calculation unit 12 uses, for example, the aforementioned equation (2) as an estimation model. In equation (2), OCV is the open circuit voltage of the battery cell 20, SOC is the charge rate of the battery cell 20, I is the value detected by the current sensor 21, and R is the internal resistance of the battery cell 20. Cap is the full charge capacity of the battery cell 20 (see equation (1) above). Δt is the execution cycle of the calculation in equation (2).

[0040] Vc is the voltage (charge-discharge polarization voltage) independent of the SOC, of ​​the polarization voltages (charge polarization voltage and discharge polarization voltage) generated by charging and discharging the battery cell 20, and p is the discharge polarization coefficient. a to d are constants set in advance based on prior experimental results and simulations.

[0041] In equation (2), the term (discharge polarization estimation term) including the discharge polarization coefficient p, constants a to d, the detected value I, the full charge capacity Cap, and the execution cycle Δt represents the voltage of the discharge polarization voltage of the battery cell 20 that depends on the change in the SOC within the execution cycle Δt. The discharge polarization estimation term calculates the change in the SOC within the execution cycle Δt based on the ratio of the integrated value of the current detected value I to the full charge capacity Cap, and estimates the increase in the discharge polarization voltage associated with the increase in diffusion resistance in the low SOC region based on the value obtained by multiplying the result by the discharge polarization coefficient p as a weighting factor. Constants a to d are determined, for example, by the diffusion resistance characteristics of the battery cell 20 during discharge, as described below.

[0042] Figure 2 This graph shows an example of how the diffusion resistance (mΩ) of a lithium-ion battery changes with SOC (%) during charge (see "△") and discharge (see "●"). The diffusion resistance during charge is essentially constant, independent of the SOC. Meanwhile, the diffusion resistance during discharge is essentially constant from 20% to 100% SOC, but increases as the SOC decreases from 0% to 20%.

[0043] As the diffusion resistance increases, the polarization voltage of the battery cell 20 increases. The discharge polarization coefficient p is a parameter that indicates the degree (weight) of the increase in the discharge polarization voltage depending on the SOC, and is larger in the low SOC range (0 to 20% in the above example) than in other ranges. By using an estimation model that includes the discharge polarization coefficient p, the state calculation unit 12 can perform highly accurate calculations, taking into account the increase in the discharge polarization voltage in the low SOC range. Note that the discharge polarization coefficient p is just one example of a coefficient.

[0044] The state calculation unit 12 obtains the detection values ​​I and V from the current sensor 21 and the voltage sensor 22, respectively, at each execution cycle Δt, and calculates CCV using equation (2). At this time, the state calculation unit 12 obtains the latest full charge capacity Cap from the capacity estimation unit 11. The state calculation unit 12 updates the internal resistance value R, the charge and discharge polarization voltage Vc, and the discharge polarization coefficient p at each execution cycle Δt based on the difference between the detection value V of the voltage sensor 22 and CCV. The state calculation unit 12 calculates the SOC and discharge polarization coefficient p, for example, using the prediction error method, so that the difference between the CCV and the detection value V of the voltage sensor is minimized.

[0045] Figure 3AThis is a functional block diagram showing an example of the state calculation unit 12. The state calculation unit 12 includes an estimation model 120 represented by equation (2), a parameter updating unit 121, and an adder 122. The estimation model 120 calculates CCV according to equation (2) based on the detection value I of the current sensor 21 at each execution cycle Δt, and outputs the calculated CCV to the adder 122.

[0046] The adder 122 receives inputs of the detection value V of the voltage sensor 22 and the CCV calculated by the estimation model 120. The detection value V of the voltage sensor 22 is a closed-circuit voltage corresponding to the current value (the detection value I of the current sensor 21) flowing through the battery cell 20. The adder 122 outputs the difference ΔV (=V-CCV) between the detection value V and the CCV to the parameter update unit 121.

[0047] SOC[t+Δt]=SOC[t]-I×Δt / 3600 / Cap×100+Ga×ΔV…(3)

[0048] R[t+Δt]=R[t]+Gb×ΔV…(4)

[0049] Vc[t+Δt]=Vc[t]+Gc×ΔV…(5)

[0050] p[t+Δt]=p[t]+Gd×ΔV…(6)

[0051] The parameter updating unit 121 updates the battery cell 20's SOC, internal resistance R, charge / discharge polarization voltage Vc, and discharge polarization coefficient p based on the difference ΔV between the detected value V and CCV, according to equations (3) to (6). The SOC, internal resistance R, charge / discharge polarization voltage Vc, and discharge polarization coefficient p are stored in the memory 14 during each calculation.

[0052] In equations (3) to (6), Ga, Gb, Gc, and Gd are gains of the differential ΔV associated with the SOC, the internal resistance value R, the charge and discharge polarization voltage Vc, and the discharge polarization coefficient p, respectively. The gains Ga to Gd are set based on prior experimental results, simulation results, etc. so as to converge the differential ΔV. It should be noted that the internal resistance value R can also be calculated based on the detection value of the temperature sensor 23 and the SOC reference map data instead of equation (4). In addition, at the beginning of the calculation, the parameter updating unit 121 sets the differential ΔV=0 based on the last retained SOC[t], and calculates SOC[t+Δt] using equation (3). It should be noted that the latest values ​​of the SOC, the internal resistance value R, the charge and discharge polarization voltage Vc, and the discharge polarization coefficient p are stored in the memory 14 at any time, and each initial value is set based on the actual measured value of the battery cell 20.

[0053] The parameter updating unit 121 adds the product of the difference ΔV and each of the gains Ga through Gd to the values ​​of the SOC, internal resistance R, charge / discharge polarization voltage Vc, and discharge polarization coefficient p at time t (see [t]). This calculates the values ​​at time (t+Δt) (see [t+Δt]). The parameter updating unit 121 outputs the SOC, internal resistance R, charge / discharge polarization voltage Vc, and discharge polarization coefficient p at time (t+Δt) to the estimation model 120.

[0054] Estimation model 120 calculates CCV based on internal resistance R, charge / discharge polarization voltage Vc, and discharge polarization coefficient p, input from parameter update unit 121, as well as detection value I from current sensor 21. At this time, state calculation unit 12 calculates OCV corresponding to SOC based on OCV map 140. Estimation model 120 outputs discharge polarization coefficient p and SOC to charge notification unit 13 via operation control unit 10.

[0055] In this manner, the state calculation unit 12 periodically calculates the CCV of the battery cell 20 using the estimation model 120 and calculates the SOC and discharge polarization coefficient p using the prediction error method so that the difference ΔV between the calculated CCV value and the detected value V by the voltage sensor 22 converges. Consequently, the state calculation unit 12 can calculate the SOC and discharge polarization coefficient p with high accuracy. It should be noted that the state calculation unit 12 may also calculate the SOC and discharge polarization coefficient p using methods other than the prediction error method.

[0056] Figure 3B This graph shows an example of the time variation of the SOC (%) during discharge of the battery cell 20. The solid line represents the SOC calculated by the estimation model 120 using the aforementioned equation (2) (Example), the dashed line represents the SOC calculated by the estimation model 120 using an equation obtained by removing the discharge polarization estimation term from equation (2) (Comparative Example), and the dotted line represents the measured SOC value.

[0057] In the case of the comparative example, no Figure 2 The increase in discharge polarization voltage due to the increase in diffusion resistance in the low SOC region shown in FIG. 1 increases the error from the measured value when the SOC is approximately 15% or less. In contrast, in the embodiment, the increase in discharge polarization voltage is calculated using the discharge polarization estimation term. Therefore, even when the SOC is approximately 15% or less, the error from the measured value is smaller than in the comparative example.

[0058] In this manner, the state calculation unit 12 can calculate the SOC with high accuracy by reflecting the increase in diffusion resistance during discharge using the discharge polarization estimation term including the discharge polarization coefficient p.

[0059] Charging notification unit processing

[0060]

[0061] Figure 4 This graph shows an example of how the discharge polarization voltage (mV) changes with SOC (%). This example shows simulation results for discharging a battery cell 20 with an SOC of 30% using the left side of equation (7) derived from equation (2) while maintaining a constant current. Here, the current magnitude (I in equation (7)) is assumed to be "large," "medium," and "small."

[0062] The greater the current flowing through the battery cell 20, the greater the discharge polarization voltage. In the SOC range of approximately 15% or less, the discharge polarization voltage increases as the SOC decreases due to an increase in diffusion resistance. At this point, the greater the current flowing through the battery cell 20, the greater the increase in the discharge polarization voltage. Therefore, in the SOC range of approximately 15% or less, the power consumed by the battery cell 20 is proportional to the increase in voltage corresponding to the SOC.

[0063] Therefore, when the SOC is low, the effect of discharge polarization varies depending on the usage of the battery cell 20. Assuming that the charging notification unit 13 always prompts the user to charge the battery cell 20 when the SOC falls below 15% regardless of the current level, when the current is "low," it is assumed that the battery cell 20 has excess power. For example, when the vehicle V is traveling on a normal road, the current flowing through the battery cell 20 is low (the current is "low"), so the charging notification is issued quickly, making it difficult to use up the battery power without waste.

[0064] On the other hand, when the current is "Large" or "Medium", the battery unit 20 is considered to be insufficiently charged. For example, when the vehicle V is traveling on a highway, the current flowing through the battery unit 20 is large (current "Large"), so the charging notification is delayed, and there is a possibility that insufficient power will hinder the vehicle V from traveling.

[0065] Therefore, the smaller the discharge polarization coefficient p is, the lower the SOC threshold (SOC threshold) used for charging notification is. For example, the charge notification unit 13 obtains a threshold corresponding to the discharge polarization coefficient p from the threshold map 141 .

[0066] Figure 5 141 is a diagram showing an example of the threshold map 141. Figure 5In the graph, the horizontal axis represents the SOC threshold (%), and the vertical axis represents the discharge polarization coefficient p. Line segment L represents the correlation between the discharge polarization coefficient p and the SOC threshold. In this example, the discharge polarization coefficient p and the SOC threshold have a linear relationship. Based on line segment L, the charge notification unit 13 obtains the SOC threshold corresponding to the discharge polarization coefficient p. For example, when the discharge polarization coefficient p is Px, SOC_TH becomes the SOC threshold according to line segment L.

[0067] The charging notification unit 13 displays a notification on the onboard display 3 urging the user to charge the battery unit 20 when the SOC is below the SOC threshold. It does not display the notification on the onboard display 3 when the SOC is greater than the SOC threshold. Specifically, the charging notification unit 13 does not issue a notification if the SOC and discharge polarization coefficient p calculated by the state calculation unit 12 are within the region Sb (on the side with the larger SOC threshold) in the two regions Sa and Sb separated by the line segment L. On the other hand, the charging notification unit 13 issues a notification if the SOC and discharge polarization coefficient p are within the region Sa (on the side with the smaller SOC threshold) in the two regions Sa and Sb separated by the line segment L.

[0068] Thus, the larger the discharge polarization coefficient p is, the larger the SOC threshold value is used by the charge notification unit 13 to determine whether or not to issue the charge notification. Figure 4 In the example shown in FIG, when the current is "large" or "medium", the determination of whether to issue a charging notification is based on an SOC threshold value that is higher than when the current is "small". Therefore, the control device 1 can urge the user to charge at an appropriate time according to the usage status of the battery unit 20.

[0069] Furthermore, the charge notification unit 13 varies the SOC threshold so that it establishes a linear relationship with the discharge polarization coefficient p. Therefore, since the diffusion resistance of the battery cell 20 during discharge increases linearly with the decrease in SOC, the SOC threshold can be set with high accuracy. It should be noted that the maximum and minimum values ​​of the SOC threshold are appropriately set according to the operating environment of the battery cell 20.

[0070] Control device operation

[0071] Figure 6 : is a flowchart showing an example of the operation of the control device 1. This operation is performed at, for example, a constant time interval.

[0072] First, the operation control unit 10 determines whether the battery cell 20 is being charged (St1). If the battery cell 20 is being charged (Yes in St1), the capacity estimation unit 11 estimates the full charge capacity Cap of the battery cell 20 (St2). If the battery cell 20 is not being charged (No in St1), the operation in St2 is not performed.

[0073] Next, the operation control unit 10 determines whether the vehicle V is traveling (St3). If the vehicle V is not traveling (No in St3), the operation ends. If the vehicle V is traveling (Yes in St3), the state calculation unit 12 calculates the SOC and discharge polarization coefficient p of the discharging battery cell 20 using the above-described method (St4). It should be noted that while the vehicle V is traveling, power is supplied from the battery cell 20 to the electric motor 5.

[0074] Next, the charge notification unit 13 calculates the SOC threshold value based on the discharge polarization coefficient p and the threshold map 141 (St5). Next, the charge notification unit 13 compares the SOC with the SOC threshold value (St6). If SOC ≤ SOC threshold value (Yes in St6), the charge notification unit 13 outputs a charge notification to the onboard display 3 (St7). If SOC > SOC threshold value (No in St6), the charge notification unit 13 does not output a charge notification and terminates its operation. The control device 1 operates in this manner.

[0075] Thus, the state calculation unit 12 calculates the SOC and discharge polarization coefficient p from the detected values ​​I and V of the current and voltage of the battery cell 20 based on the estimation model 120. The estimation model 120 estimates the discharge polarization voltage based on the SOC of the battery cell 20 and the discharge polarization coefficient p, which correlates with the amount of voltage increase in the discharge polarization voltage of the battery cell 20 corresponding to the SOC. The charging notification unit 13 notifies the user to urge charging of the battery cell 20 when the SOC is below the SOC threshold. The SOC threshold is lowered as the discharge polarization coefficient p decreases.

[0076] Therefore, the state calculation unit 12 can accurately calculate the SOC and discharge polarization coefficient p by reflecting the increase in discharge polarization voltage associated with the increase in diffusion resistance in the low SOC region during discharge of the battery cell 20. Furthermore, as the discharge polarization coefficient p decreases, the SOC threshold is lowered. This allows the charge notification unit 13 to appropriately urge the user to charge the battery based on the increase in discharge polarization voltage.

[0077] Furthermore, lithium-ion batteries are preferred for the battery unit 20 due to their high versatility. Furthermore, the battery unit 20 supplies power to the electric motor 5 that drives the vehicle V. Therefore, the control device 1 described above can be used to appropriately urge the user to charge the battery while the vehicle V is traveling. It should be noted that the control device 1 is not limited to vehicles V and can also be used, for example, to provide battery charging notifications for electronic devices such as smartphones.

[0078] The above-described embodiment is a preferred example of the present invention, but the present invention is not limited thereto and various modifications can be made without departing from the spirit of the present invention.

Claims

1. A battery management device, wherein: The battery management device has: a calculation unit that calculates an SOC and a coefficient from each measured value of a current and a voltage of a battery based on an estimation model, wherein the estimation model estimates the polarization voltage based on the SOC of the battery and the coefficient associated with an increase in the polarization voltage generated by discharge of the battery corresponding to the SOC; as well as a notification unit that, when the SOC is below a threshold, issues a notification urging the user to charge the battery; The notification unit lowers the threshold value as the coefficient becomes smaller.

2. The battery management device according to claim 1, wherein: The notification unit changes the threshold value so as to establish a linear relationship with the coefficient.

3. The battery management device according to claim 1 or 2, wherein: The calculation unit periodically calculates the voltage of the battery using the estimation model, and calculates the SOC and the coefficient so that a difference between a calculated value of the battery voltage and a measured value of the battery voltage converges.

4. The battery management device according to claim 1 or 2, wherein: The battery is a lithium-ion battery.

5. The battery management device according to claim 1 or 2, wherein: The battery supplies electric power to an electric motor that drives the vehicle.

Images