System and method for determining both an estimated battery state vector and an estimated battery parameter vector

• US 7,521,895 B2
• Filed: 03/02/2006
• Issued: 04/21/2009
• Est. Priority Date: 03/02/2006
• Status: Active Grant

• Current Assignee: LG Energy Solution Limited (LG Corporation)
• Original Assignee: LG Chem Limited (LG Corporation)
• Inventors: Plett, Gregory L.
• Primary Examiner(s): TSO, EDWARD H
• Application Number:
  US11/366,089
• Publication Number:
  US 20070236182A1
• Time in Patent Office: 1,146 Days
• Field of Search: 320/132, 320/149, 324/426, 324/427, 324/430, 324/433
• US Class Current: 320/132
• CPC Class Codes: G01R 31/367 Software therefor, e.g. for battery testing using modelling or look-up tables

Abstract

A system and a method for determining an estimated battery state vector and an estimated battery parameter vector are provided. The method includes determining an estimated battery parameter vector indicative of a parameter of the battery at the first predetermined time based on a plurality of predicted battery parameter vectors, a first plurality of predicted battery output vectors, and a battery output vector. The method further includes determining the estimated battery state vector indicative of a state of the battery at the first predetermined time based on a plurality of predicted battery state vectors, a second plurality of predicted battery output vectors, and the battery output vector.

Claims

1. A system for determining an estimated battery state vector and an estimated battery parameter vector, the estimated battery state vector being indicative of a state of a battery at a first predetermined time, the estimated battery parameter vector being indicative of a parameter of the battery at the first predetermined time, the system comprising:

a sensor configured to generate a first signal indicative of an output variable of the battery; and

a computer operably coupled to the sensor, the computer configured to determine a plurality of estimated augmented battery state vectors that are indicative of a state of the battery, a battery input noise, a sensor noise associated with a sensor measuring a battery output variable, an uncertainty of the state of the battery, an uncertainty of the battery input noise, and an uncertainty of the sensor noise at a second predetermined time prior to the first predetermined time, the computer further configured to determine a plurality of predicted battery state vectors that are indicative of the state of the battery and an uncertainty of the state of the battery at the first predetermined time based on the plurality of estimated augmented battery state vectors, the computer further configured to determine a plurality of predicted battery parameter vectors that are indicative of the parameter of the battery and an uncertainty of the parameter of the battery at the first predetermined time, the computer further configured to determine a first plurality of predicted battery output vectors that are indicative of at least one output variable of the battery and an uncertainty of the output variable at the first predetermined time based on the plurality of predicted battery parameter vectors, the computer further configured to determine a second plurality of predicted battery output vectors that are indicative of at least one output variable of the battery and the uncertainty of the output variable at the first predetermined time based on the plurality of estimated augmented battery state vectors and the plurality of predicted battery state vectors, the computer further configured to determine a battery output vector based on the first signal, the computer further configured to determine an estimated battery parameter vector indicative of the parameter of the battery at the first predetermined time based on the plurality of predicted battery parameter vectors, the first plurality of predicted battery output vectors, and the battery output vector, the computer further configured to determine an estimated battery state vector indicative of the state of the battery at the first predetermined time based on the plurality of predicted battery state vectors, the second plurality of predicted battery output vectors, and the battery output vector.

2. The system of claim 1, wherein the computer is further configured to retrieve an estimated battery state vector indicative of the state of the battery at the second predetermined time from a memory device;

the computer further configured to determine a first estimated augmented battery state vector indicative of the state of the battery, the battery input noise, and the sensor noise at the second predetermined time, based on the estimated battery state vector, an expected battery input noise vector, and an expected sensor noise vector;

the computer further configured to retrieve an estimated battery state vector covariance matrix indicative of the uncertainty of the state of the battery at the second predetermined time from the memory device;

the computer further configured to determine a first estimated augmented battery state vector covariance matrix indicative based on the estimated battery state vector covariance matrix, a covariance matrix indicative of the uncertainty of the battery input noise and a covariance matrix indicative of the uncertainty of the sensor noise; and

the computer further configured to calculate the plurality of estimated augmented battery state vectors based on the first estimated augmented battery state vector and the first estimated augmented battery state vector covariance matrix.

3. The system of claim 2, wherein the computer is further configured to set one member of the plurality of estimated augmented battery state vectors equal to the first estimated augmented battery state vector;

the computer further configured to set a first additional set of L members of the plurality of estimated augmented battery state vectors equal to the first estimated augmented battery state vector added to a constant value multiplied by respective columns extracted from a matrix square-root of the first estimated augmented battery state vector covariance matrix, where L is the length of the first estimated augmented battery state vector; and

the computer further configured to set a second additional set of L members of the plurality of estimated augmented battery state vectors equal to the constant value multiplied by respective columns extracted from the matrix square-root of the first estimated augmented battery state vector covariance matrix subtracted from the first estimated augmented battery state vector.

4. The system of claim 1, wherein the computer is further configured to retrieve an estimated battery state vector indicative of the state of the battery at the second predetermined time from a memory device;

the computer further configured to determine a first estimated augmented battery state vector indicative of the state of the battery, the battery input noise, and the sensor noise at the second predetermined time based on the estimated battery state vector, an expected battery input noise vector, and an expected sensor noise vector;

the computer further configured to retrieve an estimated battery state vector square-root covariance matrix indicative of the uncertainty of the state of the battery at the second predetermined time from the memory device;

the computer further configured to determine a first estimated augmented battery state vector square-root covariance matrix indicative of the uncertainty of the state of the battery, the battery input noise, and the sensor noise, based on the estimated battery state vector square-root covariance matrix, a covariance matrix indicative of the uncertainty of the battery input noise and a covariance matrix indicative of the uncertainty of the sensor noise; and

the computer further configured to calculate the plurality of estimated augmented battery state vectors based on the first estimated augmented battery state vector and the first estimated augmented battery state vector square-root covariance matrix.

5. The system of claim 4, wherein the computer is further configured to set one member of the plurality of estimated augmented battery state vectors equal to the first estimated augmented battery state vector;

the computer further configured to set a first additional set of L members of the plurality of estimated augmented battery state vectors equal to the first estimated augmented battery state vector added to a constant value multiplied by respective columns extracted from the estimated augmented battery state vector square-root covariance matrix, where L is the length of the estimated augmented battery state vector; and

the computer further configured to set a second additional set of L members of the plurality of estimated augmented battery state vectors equal to the constant value multiplied by respective columns extracted from the first estimated augmented battery state vector square-root covariance matrix subtracted from the first estimated augmented battery state vector.

6. The system of claim 1, wherein the computer is further configured to extract values indicative of the state of the battery and of the uncertainty of the state of the battery at the second predetermined time from the plurality of estimated augmented battery state vectors, to obtain a first plurality of estimated battery state vectors;

the computer further configured to extract values indicative of the battery input noise and an uncertainty of the battery input noise at the second predetermined time from the first plurality of estimated augmented battery state vectors, to obtain a plurality of estimated battery input noise vectors;

the computer further configured to generate a battery input vector having at least one measured value of a battery input variable at the second predetermined time;

the computer further configured to determine the plurality of predicted battery state vectors based on the first plurality of estimated battery state vectors, the first plurality of predicted battery input noise vectors, and the battery input vector.

7. The system of claim 1, wherein the computer is further configured to retrieve from a memory device an estimated battery parameter vector indicative of the parameter of the battery at the second predetermined time;

the computer further configured to retrieve an estimated battery parameter vector covariance matrix indicative of an uncertainty of the parameter of the battery at the second predetermined time from the memory device;

the computer further configured to determine an estimated battery parameter noise covariance matrix indicative of an uncertainty of a parameter noise at the second predetermined time;

the computer further configured to calculate the plurality of predicted battery parameter vectors based on the estimated battery parameter vector, the estimated battery parameter vector covariance matrix, and the estimated battery parameter noise covariance matrix.

8. The system of claim 7, wherein the computer is further configured to set one member of the plurality of predicted battery parameter vectors equal to the estimated battery parameter vector;

the computer further configured to determine a predicted battery parameter vector covariance matrix indicative of an uncertainty of the parameter of the battery at the first predetermined time based on the estimated battery parameter vector covariance matrix and the estimated battery parameter noise covariance matrix;

the computer further configured to set a first additional set of L members of the plurality of predicted battery parameter vectors equal to the first estimated battery parameter vector added to a constant value multiplied by respective columns extracted from a matrix square-root of the first predicted battery parameter vector covariance matrix, where L is the length of the first estimated battery parameter vector; and

the computer further configured to set a second additional set of L members of the plurality of predicted battery parameter vectors equal to the constant value multiplied by respective columns extracted from a matrix square-root of the first predicted battery parameter vector covariance matrix subtracted from the first estimated battery parameter vector.

9. The system of claim 1, wherein the computer is further configured to retrieve an estimated battery parameter vector indicative of the parameter of the battery at the second predetermined time from the memory device;

the computer further configured to retrieve an estimated battery parameter vector square-root covariance matrix indicative of an uncertainty of the parameter of the battery at the second predetermined time from the memory device;

the computer further configured to determine an estimated battery parameter noise square-root covariance matrix indicative of an uncertainty of a parameter noise at the second predetermined time; and

the computer further configured to calculate the plurality of predicted battery parameter vectors based on the estimated battery parameter vector, the estimated battery parameter vector square-root covariance matrix, and the estimated battery parameter noise square-root covariance matrix.

10. The system of claim 9, wherein the computer is further configured to set one member of the plurality of predicted battery parameter vectors equal to the estimated battery parameter vector;

the computer further configured to determine a predicted battery parameter vector square-root covariance matrix indicative of an uncertainty of the parameter of the battery at the first predetermined time based on the estimated battery parameter vector square-root covariance matrix and the estimated battery parameter noise covariance matrix;

the computer further configured to set a first additional set of L members of the plurality of predicted battery parameter vectors equal to the first estimated battery parameter vector added to a constant value multiplied by respective columns extracted from the first predicted battery parameter vector square-root covariance matrix, where L is the length of the first estimated battery parameter vector; and

the computer further configured to set a second additional set of L members of the plurality of predicted battery parameter vectors equal to the constant value multiplied by respective columns extracted from the first predicted battery parameter vector square-root covariance matrix subtracted from the first estimated battery parameter vector.

11. The system of claim 1, wherein the computer is further configured to retrieve an estimated battery state vector indicative of the state of the battery at the second predetermined time from a memory device;

the computer further configured to generate a battery mean sensor noise vector indicative of the average sensor noise at the first predetermined time and a battery mean input noise vector indicative of the average battery input noise at the second predetermined time;

the computer further configured to generate a first battery input vector having at least one measured value of a battery input variable at the first predetermined time;

the computer further configured to generate a second battery input vector having at least one measured value of a battery input variable at the second predetermined time;

the computer further configured to determine the first plurality of predicted battery output vectors based on the plurality of predicted battery parameter vectors, the first battery input vector, the second battery input vector, the battery mean sensor noise vector, the battery mean input noise vector, and the estimated battery state vector.

12. The system of claim 1, wherein the computer is further configured to extract values indicative of the sensor noise and of the uncertainty of the sensor noise at the second predetermined time from the plurality of estimated augmented battery state vectors, to obtain a plurality of estimated sensor noise vectors;

the computer flirt her configured to generate the first battery input vector having at least one measured value of a battery input variable at the first predetermined time;

the computer further configured to determine the second plurality of predicted battery output vectors based on the plurality of predicted battery state vectors, the first battery input vector, the plurality of estimated sensor noise vectors.

13. The system of claim 1, wherein the computer is further configured to determine a first predicted battery output vector indicative of at least one output variable of the battery at the first predetermined time based on the first plurality of predicted battery output vectors;

the computer further configured to determine a predicted battery parameter vector indicative of the parameter of the battery at the first predetermined time based on an estimated battery parameter vector indicative of the parameter of the battery at the second predetermined time;

the computer further configured to determine a parameter gain matrix based on the predicted battery parameter vector, the first predicted battery output vector, the plurality of predicted battery parameter vectors, and the plurality of predicted battery output vectors; and

the computer further configured to calculate the estimated battery parameter vector based on the battery output vector, the first predicted battery output vector, the predicted battery parameter vector, and the parameter gain matrix.

14. The system of claim 13, wherein the computer is further configured to determine the first predicted battery output vector by calculating a weighted average of the plurality of predicted battery output vectors.

15. The system of claim 13, wherein the computer is further configured to determine determine a predicted battery output vector covariance matrix based on the first predicted battery output vector and the plurality of predicted battery output vectors;

the computer further configured to determine a predicted cross-covariance matrix between a battery output vector and a battery parameter vector based on the first predicted battery output vector, the plurality of predicted battery output vectors, the predicted battery parameter vector, and the plurality of predicted battery parameter vectors; and

the computer further configured to calculate the parameter gain matrix based on the predicted cross-covariance matrix and the predicted battery output vector covariance matrix.

16. The system of claim 15, wherein the computer is further configured to determine a predicted battery output vector square-root covariance matrix based on the first predicted battery output vector and the plurality of predicted battery output vectors;

the computer further configured to calculate the predicted battery output vector covariance matrix based on the predicted battery output vector square-root covariance matrix and the predicted battery output vector square-root covariance matrix.

17. The system of claim 13, wherein the computer is further configured to determine an innovation vector based on the battery output vector and the first predicted battery output vector;

the computer further configured to determine an update vector based on the parameter gain matrix and the innovation vector; and

the computer further configured to calculate the estimated battery parameter vector based on the predicted battery parameter vector and the update vector.

18. The system of claim 13, wherein the computer is further configured to determine an estimated battery parameter vector covariance matrix at the first predetermined time based on the first predicted battery parameter vector, the first plurality of predicted battery parameter vectors, the parameter gain matrix, and the first predicted battery output vector covariance matrix.

19. The system of claim 13, wherein the computer is further configured to determine an estimated battery parameter vector square-root covariance matrix at the first predetermined time based on the first predicted battery parameter vector, the first plurality of predicted battery parameter vectors, the parameter gain matrix, and a predicted battery output vector square-root covariance matrix.

20. The system of claim 1, wherein the computer is further configured to determine a predicted battery state vector indicative of the state of the battery at the first predetermined time based on the plurality of predicted battery state vectors;

the computer further configured to determine a second predicted battery output vector indicative of at least one output variable of the battery at the first predetermined time based on the plurality of predicted output vectors;

the computer further configured to determine a state gain matrix based on the predicted battery state vector, the second predicted battery output vector, the plurality of predicted battery state vectors, and the second plurality of predicted battery output vectors; and

the computer further configured to calculate the estimated battery state vector based on the predicted battery state vector, the second predicted battery output vector, the state gain matrix, and the battery output vector.

21. The system of claim 20, wherein the computer is configured to determine the predicted battery state vector by calculating a weighted average of the plurality of predicted battery state vectors.

22. The system of claim 20, wherein the computer is configured to determine the second predicted battery output vector by calculating a weighted average of the second plurality of predicted battery output vectors.

23. The system of claim 20, wherein the computer is further configured to determine a predicted battery output vector covariance matrix based on the second predicted battery output vector and the second plurality of predicted battery output vectors;

the computer further configured to determine a predicted cross-covariance matrix between a battery output vector and a battery state vector based on the second predicted battery output vector, the second plurality of predicted battery output vectors, the predicted battery state vector, and the plurality of predicted battery state vectors; and

the computer further configured to calculate the state gain matrix based on the predicted cross-covariance matrix and the predicted battery output vector covariance matrix.

24. The system of claim 23, wherein the computer is further configured to determine a predicted battery output vector square-root covariance matrix based on the second predicted battery output vector and the second plurality of predicted battery output vectors;

and

the computer further configured to calculate the predicted battery output vector covariance matrix based on the predicted battery output vector square-root covariance matrix and the predicted battery output vector square-root covariance matrix.

25. The system of claim 20, wherein the computer is further configured to determine an innovation vector based on the second battery output vector and the second predicted battery output vector;

the computer further configured to determine an update vector based on the state gain matrix and the innovation vector; and

the computer further configured to calculate the estimated battery state vector based on the predicted battery state vector and the update vector.

26. The system of claim 20, wherein the computer is further configured to determine an estimated battery state vector covariance matrix at the first predetermined time based on the predicted battery state vector, the plurality of predicted battery state vectors, the state gain matrix, and the predicted battery output vector covariance matrix.

27. The system of claim 20, wherein the computer is further configured to determine an estimated battery state vector square-root covariance matrix at the first predetermined time based on the predicted battery state vector, the plurality of predicted battery state vectors, the state gain matrix, and a predicted battery output vector square-root covariance matrix.

28. A method for determining an estimated battery state vector and an estimated battery parameter vector, the estimated battery state vector being indicative of a state of a battery at a first predetermined time, the estimated battery parameter vector being indicative of a parameter of the battery at the first predetermined time, the method comprising:

determining a plurality of estimated augmented battery state vectors that are indicative of a state of the battery, a battery input noise, a sensor noise associated with a sensor measuring a battery output variable, an uncertainty of the state of the battery, an uncertainty of the battery input noise, and an uncertainty of the sensor noise at a second predetermined time prior to the first predetermined time;

determining a plurality of predicted battery state vectors that are indicative of the state of the battery and an uncertainty of the state of the battery at the first predetermined time based on the plurality of estimated augmented battery state vectors;

determining a plurality of predicted battery parameter vectors that are indicative of the parameter of the battery and an uncertainty of the parameter of the battery at the first predetermined time;

determining a first plurality of predicted battery output vectors that are indicative of at least one output variable of the battery and an uncertainty of the output variable at the first predetermined time based on the plurality of predicted battery parameter vectors;

determining a second plurality of predicted battery output vectors that are indicative of at least one output variable of the battery and the uncertainty of the output variable at the first predetermined time based on the plurality of estimated augmented battery state vectors and the plurality of predicted battery state vectors;

determining a battery output vector having at least one measured value of a battery output variable obtained at the first predetermined time;

determining an estimated battery parameter vector indicative of the parameter of the battery at the first predetermined time based on the plurality of predicted battery parameter vectors, the first plurality of predicted battery output vectors, and the battery output vector;

determining an estimated battery state vector indicative of the state of the battery at the first predetermined time based on the plurality of predicted battery state vectors, the second plurality of predicted battery output vectors, and the battery output vector; and

storing the estimated battery state vector and the estimated battery parameter vector in a memory device.