JP2013246088A - Method and device for estimating internal resistance of battery - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and a device for estimating internal resistance R of a battery, which can reduce an error included in the estimated internal resistance R and can improve estimation accuracy of the internal resistance R.SOLUTION: In a method and a device for estimating internal resistance R of a battery 1, a current value I and a voltage value CCV of the battery 1 are monitored, and data D is accumulated including a current variation amount ΔI as a variation amount of the current value I when the current value I is changed by a predetermined threshold value Th or more, and a voltage variation amount ΔV as a variation amount of the voltage value CCV. On the basis of the plurality of accumulated data D, a gradient ΔV/ΔI of a regression line, which represents the voltage variation amount ΔV by using the current variation amount ΔI as a variable, is obtained to estimate the gradient ΔV/ΔI as the internal resistance R of the battery.

Description

  The present invention relates to a battery internal resistance estimation method and apparatus for estimating battery internal resistance.

Generally, the relationship between the voltage (CCV: Closed Circuit Voltage) at the time of battery discharge and the open circuit voltage (OCV) is expressed as follows. Here, I is the current value of the battery, and R is the internal resistance of the battery.
OCV = CCV + I × R + polarization The internal resistance R of the battery is generated without delay with respect to the change in the current value. On the other hand, it is known that the influence of polarization occurs with a delay of about 0.1 seconds with respect to the current change and saturates at a constant time.

  For example, the following Patent Document 1 proposes a method for calculating the internal resistance of a battery using the above relationship. That is, in the conventional method, the secondary battery is pulse-discharged, the voltage change ΔV between the voltage value before the pulse discharge and the voltage value when the pulse discharge is performed, and the current value and the pulse before the pulse discharge are performed. A current change amount ΔI with the current value at the time of discharging is obtained. Then, the internal resistance R of the battery is estimated by calculating the voltage change amount ΔV / current change amount ΔI according to Ohm's law.

JP 2000-121710 A

  In the conventional method and apparatus for estimating the internal resistance of a battery as described above, the internal resistance R of the battery is estimated based on the voltage change amount ΔV and the current change amount ΔI when a single pulse discharge is performed. Therefore, when a sudden error is included in the voltage or current measurement, the error is reflected as it is in the estimated internal resistance R.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to reduce the error included in the estimated internal resistance R and to improve the estimation accuracy of the internal resistance R. It is to provide a resistance estimation method and apparatus.

  The battery internal resistance estimation method according to the present invention monitors the battery current value I and the voltage value CCV, and the current change amount that is the change amount of the current value I when the current value I changes by a predetermined threshold Th or more. Data D of ΔI and voltage change amount ΔV, which is the change amount of voltage value CCV, is accumulated. Based on the accumulated data D, the slope ΔV of the regression line representing the voltage change amount ΔV using current change amount ΔI as a variable. / ΔI is obtained, and the slope ΔV / ΔI is estimated as the internal resistance R of the battery.

  The battery internal resistance estimation device according to the present invention monitors the battery current value I and the voltage value CCV, and the current change amount that is a change amount of the current value I when the current value I changes by a predetermined threshold Th or more. Based on a plurality of data D accumulated by ΔI and a voltage change amount ΔV that is a change amount of the voltage value CCV, and a plurality of data D accumulated by the accumulation unit, the voltage change amount ΔV is set with the current change amount ΔI as a variable An estimation unit that obtains the slope ΔV / ΔI of the regression line to be expressed and estimates the slope ΔV / ΔI as the internal resistance R of the battery;

  According to the battery internal resistance estimation method and apparatus of the present invention, the regression line slope ΔV / ΔI representing the voltage change ΔV is obtained based on the accumulated data D and the current change ΔI as a variable, Since the slope ΔV / ΔI is estimated as the internal resistance R of the battery, the error included in the estimated internal resistance R can be reduced, and the estimation accuracy of the internal resistance R can be improved.

It is explanatory drawing which shows the internal resistance estimation apparatus of the battery by Embodiment 1 of this invention. It is explanatory drawing which shows the basic concept of the estimation method of the internal resistance R based on the some data D accumulate | stored in the accumulation | storage part of FIG. It is explanatory drawing which shows the specific application of the estimation method of FIG. It is a flowchart which shows the operation | movement as the whole internal resistance estimation apparatus of FIG.

Hereinafter, embodiments for carrying out the present invention will be described with reference to the drawings.
Embodiment 1 FIG.
FIG. 1 is an explanatory diagram showing a battery internal resistance estimation device 3 according to Embodiment 1 of the present invention. In the figure, a battery 1 is, for example, a battery cell constituting an assembled battery mounted on a vehicle, and supplies power to a load 20 and is charged by power from a power supply device 21. The load 20 is, for example, an in-vehicle motor, and the power supply device 21 is, for example, an in-vehicle generator or a charger that supplies power from the external power source to the battery 1.

  Connected to the battery 1 is an internal resistance estimation device 3 configured by, for example, a microcomputer or the like that performs the internal resistance estimation method of the battery 1 of the present embodiment. The internal resistance estimation device 3 includes an accumulation unit 30 and an estimation unit 31.

  The storage unit 30 monitors the current value I and the voltage value CCV of the battery 1 in use. Specifically, the storage unit 30 samples the current value I and the voltage value CCV at short intervals of 0.1 seconds or less. The reason why the sampling interval is set to 0.1 seconds or less is that the voltage value CCV in which the change in the influence of the polarization accompanying the current change is extremely small can be detected by sampling the voltage value CCV at such an interval. The battery 1 in use is the battery 1 when power is being supplied to the load 20 (during discharging) and when power is being supplied from the power supply device 21 (during charging). The charging / discharging of the battery 1 is not limited to actual use, and may be intentionally performed by controlling the load 20 and the like.

Further, when the sampled current value I changes by a predetermined threshold Th or more, for example, about 10 A, the storage unit 30 uses the current change amount ΔI that is the change amount of the current value I and the change amount of the voltage value CCV. A certain voltage change amount ΔV is calculated. To be more specific, the storage unit 30, when the difference between the n-th current value I _n and the (n + 1) th current value I _{n + 1} sampled is equal to or larger than the threshold Th, the n-th current value I _n and calculates the current change amount [Delta] I _n is the difference between the (n + 1) th current value _{I n + 1,} n-th voltage value CCV _n and the (n + 1) th voltage _{CCV n + 1} and the voltage change amount ΔV is the difference _n is calculated. Note that n is an arbitrary integer of 1 or more.

If, when the time of sampling the n-th voltage value CCV _n is after elapse of a predetermined time since start of use of the battery 1, the voltage drop due to the polarization in the n-th voltage value CCV _n (polarization component) Already included. However, by shortening the sampling interval as described above, the n polarization component included in the n-th voltage value CCV _n is the n + 1 polarization component and substantially contained in the (n + 1) th voltage CCV _n Can be considered equal. Therefore, cancel the n polarization component and a second n + 1 polarization component from each other when calculating the voltage change amount [Delta] V _n, the voltage variation [Delta] V _n does not include polarization components.

  Further, the storage unit 30 stores data D of the calculated current change amount ΔI and voltage change amount ΔV. That is, the storage unit 30 stores data D of a plurality of current change amounts ΔI and voltage change amounts ΔV when the current value I changes by a threshold Th or more.

Furthermore, when accumulating the data D, the accumulating unit 30 accumulates the current change amount ΔI and the voltage change amount ΔV in association with the temperature of the battery 1 and the SOC (State of Charge). To be more specific, the storage unit 30, when storing the data D according to the current change amount [Delta] I _n and the voltage variation [Delta] V _n as described above, the n-th current value I _n and the voltage value CCV _n or the The temperature and SOC of the battery 1 when sampling the (n + 1) th current value I _{n + 1} and voltage value CCV _{n + 1} are associated with the current change amount ΔI _n and the voltage change amount ΔV _n . The temperature is detected by a temperature sensor (not shown). The SOC is estimated based on, for example, the open circuit voltage OCV of the battery 1 or based on the integrated value of the current output from the battery 1.

  The estimation unit 31 estimates the internal resistance R of the battery 1 based on the plurality of data D stored in the storage unit 30. The estimated internal resistance R is used to determine, for example, the deterioration state of the battery 1.

  Next, FIG. 2 is an explanatory diagram showing a basic concept of a method for estimating the internal resistance R based on a plurality of data D stored in the storage unit 30 of FIG. The internal resistance R can be obtained by dividing the voltage change amount ΔV by the current change amount ΔI. That is, the internal resistance R is equivalent to the slope ΔV / ΔI of the regression line f (ΔI) representing the voltage change amount ΔV with the current change amount ΔI as a variable.

  Based on a plurality of data D stored in the storage unit 30, the estimation unit 31 uses the current change amount ΔI as a variable and a slope ΔV / ΔI of a regression line f (ΔI) representing the voltage change amount ΔV as shown in FIG. And the slope ΔV / ΔI is estimated as the internal resistance R of the battery 1. The slope ΔV / ΔI of the regression line f (ΔI) can be obtained by performing regression analysis such as a least square method on the data D. In other words, the slope ΔV / ΔI of the regression line f (ΔI) described above is the slope of the linear function that most closely approximates the change characteristic between the voltage change amount ΔV and the current change amount ΔI of the battery 1 represented by the data D. It is.

  Further, when the estimation unit 31 obtains the slope ΔV / ΔI, the estimation unit 31 uses a condition that the voltage change amount ΔV is also zero when the current change amount ΔI is zero. In other words, the condition that the ΔV intercept of the regression line f (ΔI) is 0 is used. By using such conditions, the slope ΔV / ΔI can be obtained more easily.

Furthermore, the estimation unit 31 weights each data D stored in the storage unit 30 according to the magnitude of the current change amount ΔI before obtaining the slope ΔV / ΔI. For example, when the straight line that minimizes the square of the error with each data D is the regression line f (ΔI), the slope ΔV / ΔI of the regression line f (ΔI) is obtained by multiplying each error by a weighting coefficient. Ask. The weighting coefficient is a coefficient that sets the weight of the i-th data as ΔI _i / ΔI _max with reference to the largest maximum current change amount ΔI _max among the current change amounts ΔI stored in the storage unit 30. i is an arbitrary positive integer. The reason why such weighting is performed is because the ratio of the voltage drop due to the internal resistance R is larger as the current change amount ΔI is larger, and this is reflected in the calculation of the slope ΔV / ΔI.

Next, FIG. 3 is an explanatory diagram showing a specific application of the estimation method of FIG. As described above, the storage unit 30 stores the temperature D and the SOC of the battery 1 in association with the data D of the current change amount ΔI and the voltage change amount ΔV. As shown in FIG. 3, the estimating unit 31 obtains the slope ΔV / ΔI of the regression line f (ΔI) for each SOC range of the battery 1 and for each temperature range of the battery 1, and the battery 1 for each of these ranges. The internal resistances _{R1-1 ... X-Y} are estimated. The reason why the internal resistances R _{1-1... XY} are estimated for each such range is that the internal resistances R _1-1 . In the present embodiment, the inclination ΔV / ΔI is described for each SOC range and each temperature range. However, the inclination ΔV / Δ for each one of the SOC range and the temperature range is described. ΔI may be obtained.

  Next, FIG. 4 is a flowchart showing the overall operation of the internal resistance estimation device 3 of FIG. In the figure, when the internal resistance estimation device 3 is powered on, the storage unit 30 samples the current value I and the voltage value CCV of the battery 1 (step S1). Next, noise filtering is performed on the sampled current value I and voltage value CCV (step S2), and the current change amount ΔI and the voltage change amount ΔV when the current value I changes by a predetermined threshold Th or more. Is stored by the storage unit 30 (step S3). At this time, the SOC and temperature of the battery 1 corresponding to the current change amount ΔI and the voltage change amount ΔV are associated with each other.

  Next, the estimation unit 31 determines whether or not the number of data D accumulated by the accumulation unit 30 is greater than or equal to a certain value (step S4). The determination as to whether or not the number of data D is greater than or equal to a certain value is made for each SOC range and each temperature range.

  If it is determined at this determination that the number of data D is not equal to or greater than a certain value, the storage operation (steps S1 to S3) of data D by the storage unit 30 is continued. On the other hand, when it is determined that the number of data D is greater than or equal to a certain value, the estimation unit 31 calculates the slope ΔV / ΔI of the regression line f (ΔI) representing the voltage change amount ΔV using the current change amount ΔI as a variable. It is obtained for each SOC range and for each temperature range (step S5), and the inclination ΔV / ΔI is estimated to be the internal resistance R of the battery 1 in each range (step S6).

  In such an internal resistance estimation method for the battery 1 and the device 3 thereof, the voltage change amount ΔV is expressed using the current change amount ΔI as a variable based on the accumulated data D of the plurality of current change amounts ΔI and the voltage change amount ΔV. Since the slope ΔV / ΔI of the regression line is obtained and this slope ΔV / ΔI is estimated as the internal resistance R of the battery 1, even if a sudden error is included in one data D, the estimated internal The influence of error on the resistance R can be reduced. That is, the error included in the estimated internal resistance R can be reduced, and the estimation accuracy of the internal resistance R can be improved.

  Further, when storing the data D, the battery SOC is stored in association with the current change amount ΔI and the voltage change amount ΔV, and the slope ΔV / ΔI is obtained for each SOC range. The resistance R can be estimated more finely, and the state of the battery 1 can be grasped more accurately.

  Further, when storing the data D, the battery temperature is stored in association with the current change amount ΔI and the voltage change amount ΔV, and the slope ΔV / ΔI is obtained for each temperature range. The resistance R can be estimated more finely, and the state of the battery 1 can be grasped more accurately.

  Furthermore, since the condition that the voltage change amount ΔV is 0 when the current change amount ΔI is 0 is used when obtaining the slope ΔV / ΔI, the slope ΔV / ΔI can be obtained more easily.

  In addition, since the accumulated data D is weighted according to the magnitude of the current change amount ΔI before the slope ΔV / ΔI is obtained, the voltage drop due to the internal resistance R due to the magnitude of the current change amount ΔI is reduced. The ratio can be reflected in the calculation of the slope ΔV / ΔI, and the slope ΔV / ΔI can be calculated more accurately.

DESCRIPTION OF SYMBOLS 1 Battery 3 Internal resistance estimation apparatus 30 Accumulation part 31 Estimation part

Claims (6)

The battery current value I and the voltage value CCV are monitored, and the current change amount ΔI, which is the change amount of the current value I when the current value I changes by a predetermined threshold Th or more, and the change amount of the voltage value CCV. Accumulating data D with a certain voltage change amount ΔV,
Based on the accumulated data D, a slope ΔV / ΔI of a regression line representing the voltage change amount ΔV is obtained using the current change amount ΔI as a variable, and the slope ΔV / ΔI is determined as the internal resistance R of the battery. An internal resistance estimation method for a battery, characterized by comprising: When storing the data D, the SOC of the battery is stored in association with the current change amount ΔI and the voltage change amount ΔV,
The battery internal resistance estimation method according to claim 1, wherein the slope ΔV / ΔI is obtained for each SOC range. When accumulating the data D, the current change amount ΔI and the voltage change amount ΔV are accumulated in association with the battery temperature,
The battery internal resistance estimation method according to claim 1, wherein the slope ΔV / ΔI is obtained for each temperature range. The condition that the voltage change amount ΔV is also 0 when the current change amount ΔI is 0 is used when obtaining the slope ΔV / ΔI. The battery internal resistance estimation method according to claim 1. 5. The weight of the accumulated data D is weighted according to the magnitude of the current change amount ΔI before the slope ΔV / ΔI is obtained. 5. The battery internal resistance estimation method according to claim 1. The battery current value I and the voltage value CCV are monitored, and the current change amount ΔI, which is the change amount of the current value I when the current value I changes by a predetermined threshold Th or more, and the change amount of the voltage value CCV. An accumulation unit for accumulating data D with a certain voltage change amount ΔV;
Based on the plurality of data D accumulated by the accumulating unit, a slope ΔV / ΔI of a regression line representing the voltage change amount ΔV is obtained using the current change amount ΔI as a variable, and the slope ΔV / ΔI is obtained from the battery. An internal resistance estimation device for a battery, comprising: an estimation unit that estimates an internal resistance R.

Images