Methods for on-line, high-accuracy estimation of battery state of power

• US 9,989,595 B1
• Filed: 12/30/2014
• Issued: 06/05/2018
• Est. Priority Date: 12/31/2013
• Status: Active Grant

• Current Assignee: HRL Laboratories LLC (The Boeing Co.)
• Original Assignee: HRL Laboratories LLC (The Boeing Co.)
• Inventors: Wang, Shuoqin, Vu, Luan D., Jiang, Qin
• Primary Examiner(s): Ngon, Ricky
• Application Number:
  US14/586,828
• Time in Patent Office: 1,253 Days
• Field of Search: 702 63
• CPC Class Codes: G01R 31/3648 comprising digital calculation means, e.g. for performing an algorithm
• CPC Class Codes: G01R 31/382 Arrangements for monitoring battery or accumulator variables, e.g. SoC
• CPC Class Codes: G01R 31/387 Determining ampere-hour charge capacity or SoC

Abstract

Some variations provide a method for real-time estimation of state of charge and state of power of a battery, comprising: (a) cycling a battery with a driving profile; (b) utilizing a recursive algorithm that relates battery terminal voltage to battery current, wherein the algorithm includes open-circuit voltage and a finite-impulse-response filter to dynamically model kinetic voltage; measuring the battery terminal voltage and the battery current at least at a first time and a second time during cycling; calculating battery open-circuit voltage and finite-impulse-response filter parameters; calculating battery state of charge based on the open-circuit voltage; and calculating battery state of power based on the open-circuit voltage and the finite-impulse-response filter parameters. An extended Kalman filtering technique is incorporated for real-time updating of FIR model parameters. Only a single FIR filter is necessary, making these methods applicable for battery-powered systems with limited computing and storage capabilities.

Claims

1. A method for real-time estimation of state of charge and state of power of a battery, said method comprising:

(a) cycling a battery with a driving profile;

(b) initializing a recursive algorithm that relates battery terminal voltage to battery current, wherein said recursive algorithm includes voltage components of (i) open-circuit voltage and (ii) a finite-impulse-response filter to dynamically model kinetic voltage;

(c) measuring said battery terminal voltage and said battery current at least at a first time and a second time during said cycling;

(d) calculating, using said recursive algorithm, battery open-circuit voltage and finite-impulse-response filter parameters;

(e) calculating battery state of charge based on said open-circuit voltage from step (d), using a look-up table, graph, equation, or combination thereof;

(f) calculating battery state of power based on said open-circuit voltage and said finite-impulse-response filter parameters from step (d); and

(g) managing said battery by adjusting electrical current and/or voltage from or to said battery in response to said battery state of charge and said battery state of power, to dynamically regulate said battery.

2. The method of claim 1, wherein said driving profile includes a plurality of double-pulse sequences, wherein each of said double-pulse sequences comprises a discharge pulse having a pulse width and a pulse amplitude, a charge pulse having said pulse width and said pulse amplitude, and a zero-current period.

3. The method of claim 1, wherein said cycling in step (a) is performed for at least 1 minute.

4. The method of claim 1, wherein said initializing in step (b) uses said battery terminal voltage measured at the beginning of said cycling.

5. The method of claim 1, wherein said kinetic voltage is formulated as the convolution of said finite-impulse-response filter parameters with said battery current.

6. The method of claim 1, said method comprising adjusting the number of said finite-impulse-response filter parameters to improve stability of said method or to accommodate battery kinetics.

7. The method of claim 1, wherein said recursive algorithm utilizes an extended Kalman filter.

8. The method of claim 7, wherein said extended Kalman filter is the only filter in said recursive algorithm.

9. The method of claim 1, said method further comprising generating said look-up table, graph, equation, or combination thereof to form a pre-determined correlation of said open-circuit voltage with said battery state of charge.

10. The method of claim 1, wherein said state of power is calculated as a product of a time-varying voltage and a constant current.

11. The method of claim 10, wherein said constant current is the maximum positive (charging) or negative (discharging) current applicable to said battery.

12. The method of claim 1, wherein said state of power is calculated as a product of a constant voltage and a time-varying current.

13. The method of claim 12, wherein said constant voltage is the maximum voltage (charging) or minimum voltage (discharging) current applicable to said battery.

14. A system for dynamically characterizing the state of charge and state of power a battery, said system comprising a battery and a programmable power-supply apparatus electrically linked with said battery, wherein said programmable power-supply apparatus is programmed using non-transitory memory with executable code for executing the steps of:

(a) cycling a battery with a driving profile;

(b) initializing a recursive algorithm that relates battery terminal voltage to battery current, wherein said recursive algorithm includes voltage components of (i) open-circuit voltage and (ii) a finite-impulse-response filter to dynamically model kinetic voltage;

(c) measuring said battery terminal voltage and said battery current at least at a first time and a second time during said cycling;

(d) calculating, using said recursive algorithm, battery open-circuit voltage and finite-impulse-response filter parameters;

(e) calculating battery state of charge based on said open-circuit voltage from step (d), using a look-up table, graph, equation, or combination thereof;

(f) calculating battery state of power based on said open-circuit voltage and said finite-impulse-response filter parameters from step (d); and

(g) managing said battery by adjusting electrical current and/or voltage from or to said battery in response to said battery state of charge and said battery state of power, to dynamically regulate said battery.

15. The system of claim 14, wherein said driving profile includes a plurality of double-pulse sequences, wherein each of said double-pulse sequences comprises a discharge pulse having a pulse width and a pulse amplitude, a charge pulse having said pulse width and said pulse amplitude, and a zero-current period.

16. The system of claim 14, wherein said kinetic voltage is formulated as the convolution of said finite-impulse-response filter parameters with said battery current.

17. The system of claim 15, wherein said recursive algorithm utilizes an extended Kalman filter.

18. The system of claim 14, wherein said state of power is calculated as a product of a time-varying voltage and a constant current, wherein said constant current is the maximum positive (charging) or negative (discharging) current applicable to said battery.

19. The system of claim 14, wherein said state of power is calculated as a product of a constant voltage and a time-varying current, wherein said constant voltage is the maximum voltage (charging) or minimum voltage (discharging) current applicable to said battery.

20. A non-transitory computer-readable medium containing computer instructions stored therein for causing a computer processor to perform steps of:

(a) cycling a battery with a driving profile;

(b) initializing a recursive algorithm that relates battery terminal voltage to battery current, wherein said recursive algorithm includes voltage components of (i) open-circuit voltage and (ii) a finite-impulse-response filter to dynamically model kinetic voltage;

(c) measuring said battery terminal voltage and said battery current at least at a first time and a second time during said cycling;

(d) calculating, using said recursive algorithm, battery open-circuit voltage and finite-impulse-response filter parameters;

(e) optionally adjusting the number of said finite-impulse-response filter parameters to improve stability of said method or to accommodate battery kinetics;

(f) calculating battery state of charge based on said open-circuit voltage from step (d), using a look-up table, graph, equation, or combination thereof;

(g) calculating battery state of power based on said open-circuit voltage and said finite-impulse-response filter parameters from step (d); and

(h) managing said battery by adjusting electrical current and/or voltage from or to said battery in response to said battery state of charge and said battery state of power, to dynamically regulate said battery.