METHOD AND SYSTEM FOR ITERATIVELY DETERMINING STATE OF CHARGE OF A BATTERY CELL

• US 20160202325A1
• Filed: 01/14/2016
• Published: 07/14/2016
• Est. Priority Date: 01/14/2015
• Status: Active Grant

• Current Assignee: Corvus Energy, Inc.
• Original Assignee: Corvus Energy Ltd.
• Inventors: Tang, Siu Lun Isaac
• Granted Patent: US 10,422,834 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G01R 31/3648 comprising digital calculation means, e.g. for performing an algorithm
• CPC Class Codes: G01R 31/367 Software therefor, e.g. for battery testing using modelling or look-up tables
• CPC Class Codes: G01R 31/382 Arrangements for monitoring battery or accumulator variables, e.g. SoC
• CPC Class Codes: G01R 31/3842 combining voltage and current measurements
• CPC Class Codes: G01R 31/396 Acquisition or processing of data for testing or for monitoring individual cells or groups of cells within a battery

Abstract

A method and system for iteratively determining state of charge (SOC) of a battery cell (“selected cell”) using a controller. For each of the iterations, the controller determines determine a predicted SOC of the selected cell; determines a predicted error covariance of the predicted SOC; and updates the predicted SOC and the predicted error covariance for use in subsequent iterations of the method. The updated SOC is treated as the SOC of the selected cell for that iteration of the method. Cell voltage used to determine the updated SOC may be low-pass filtered prior to its use. When the selected cell is one of multiple cells in a battery pack, the selected cell may be selected to have the lowest SEV of the cells in the pack. The error covariance may also vary directly with pack current magnitude to model inaccuracies that generally directly vary with current flow.

Claims

1. A method for iteratively determining state of charge (SOC) of a battery cell (“

selected cell”

), the method comprising using a controller to perform a number of iterations, wherein each of the iterations comprises using the controller to;

a) determine a predicted SOC of the selected cell from an initial SOC value or an SOC of the selected cell determined from performing a previous one of the iterations;

b) determine a predicted error covariance of the predicted SOC from an initial error covariance value or an error covariance determined from performing the previous one of the iterations;

c) obtain a measurement of current flowing through the selected cell (“

current measurement”

);

d) update the predicted SOC to determine an updated SOC of the selected cell, wherein updating the predicted SOC comprises adding a weighted correction factor that is determined using the predicted error covariance to the predicted SOC and wherein the weighted correction factor decreases as the magnitude of the current measurement increases and increases as the magnitude of the current measurement decreases; and

e) update the predicted error covariance to determine an updated error covariance.

2. The method of claim 1 wherein the selected cell is selected from multiple battery cells, the weighted correction factor is determined using a difference between a measured cell voltage (SEV) and a predicted SEV of the selected cell, and the method further comprises using the controller to determine the selected cell to be the battery cell having the lowest measured SEV.

3. The method of claim 2 wherein determining the selected cell to be the battery cell having the lowest measured SEV comprises using the controller to:

a) obtain, multiple times during an election period, measurements of the SEV of each of the battery cells; and

b) determine the selected cell to be the battery cell found most often to have the lowest measured SEV during the election period.

4. The method of claim 3 wherein the controller repeatedly determines, for each of at least two of the election period, which of the battery cells most often has the lowest measured SEV, and wherein the selected cell is determined from measurements obtained during the election period that has most recently elapsed.

5. The method of claim 2 wherein each of the iterations further comprises using the controller to low-pass filter the measured SEV prior to using it to determine the weighted correction factor.

6. The method of claim 5 wherein an exponentially-weighted infinite impulse response filter is used to low-pass filter the measured SEV.

7. The method of claim 6 wherein the updated error covariance is used to determine the predicted error covariance of a subsequent one of the iterations, wherein

8. The method of claim 7 wherein the measurement covariance is of the form A·

I+b, where A and b are constants and I is the magnitude of the current measurement.

9. The method of claim 7 further comprising using the controller to low-pass filter the measurement covariance prior to using it to determine the updated error covariance.

10. The method of claim 9 wherein an exponentially-weighted infinite impulse response filter is used to low-pass filter the measurement covariance.

11. A system for iteratively determining state of charge (SOC) of a battery cell (“

selected cell”

), wherein the selected cell comprises one of multiple battery cells, the system comprising;

a) voltmeters for measuring a voltage across each of the battery cells;

b) an ammeter for measuring a current flowing through the battery cell; and

c) a controller communicatively coupled to the voltmeters and the ammeter, the controller configured to perform a method comprising using the controller to perform a number of iterations, wherein each of the iterations comprises using the controller to;

i) determine a predicted SOC of the selected cell from an initial SOC value or an SOC of the selected cell determined from performing a previous one of the iterations;

ii) determine a predicted error covariance of the predicted SOC from an initial error covariance value or an error covariance determined from performing the previous one of the iterations;

iii) obtain a measurement of current flowing through the selected cell (“

current measurement”

);

iv) update the predicted SOC to determine an updated SOC of the selected cell, wherein updating the predicted SOC comprises adding a weighted correction factor that is determined using the predicted error covariance to the predicted SOC and wherein the weighted correction factor decreases as the magnitude of the current measurement increases and increases as the magnitude of the current measurement decreases; and

v) update the predicted error covariance to determine an updated error covariance.

12. The system of claim 11 wherein the selected cell is selected from multiple battery cells, the weighted correction factor is determined using a difference between a measured cell voltage (SEV) and a predicted SEV of the selected cell, and the method further comprises using the controller to determine the selected cell to be the battery cell having the lowest measured SEV.

13. The system of claim 12 wherein determining the selected cell to be the battery cell having the lowest measured SEV comprises using the controller to:

a) obtain, multiple times during an election period, measurements of the SEV of each of the battery cells; and

b) determine the selected cell to be the battery cell found most often to have the lowest measured SEV during the election period.

14. The system of claim 13 wherein the controller repeatedly determines, for each of at least two of the election period, which of the battery cells most often has the lowest measured SEV, and wherein the selected cell is determined from measurements obtained during the election period that has most recently elapsed.

15. The system of claim 12 wherein each of the iterations further comprises using the controller to low-pass filter the measured SEV prior to using it to determine the weighted correction factor.

16. The system of claim 15 wherein an exponentially-weighted infinite impulse response filter is used to low-pass filter the measured SEV.

17. The system of claim 16 wherein the updated error covariance is used to determine the predicted error covariance of a subsequent one of the iterations, wherein

18. The system of claim 17 wherein the measurement covariance is of the form A·

I+b, where A and b are constants and I is the magnitude of the current measurement.

19. The system of claim 17 further comprising using the controller to low-pass filter the measurement covariance prior to using it to determine the updated error covariance.

20. The system of claim 19 wherein an exponentially-weighted infinite impulse response filter is used to low-pass filter the measurement covariance.