IP Library Granted Patent US 12693339
Granted Patent B2
US 12693339 · App. 18/307,327 · Granted Jul 28, 2026

Estimation of state of charge of a battery assembly

Inventors: Yue-Yun Wang (Troy, MI); Chengwu Duan (Shanghai, CN); Chen-fang Chang (Bloomfield Hills, MI)
Assignee: GM GLOBAL TECHNOLOGY OPERATIONS LLC
G01R31/382B60L50/64B60L58/12G01R31/3648G01R31/367G01R31/392
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12693339
App. No.
18/307,327
Granted
Jul 28, 2026
Kind
B2
Abstract

A system for estimating a parameter of a battery assembly includes a sensing device connected to a cell of the battery assembly, the battery assembly having a first battery chemistry, the sensing device including a measurement cell having a second battery chemistry that is different than the first battery chemistry. The system includes an estimator configured to acquire a set of first estimates of a first state of charge (SOC) of the battery assembly, acquire a set of second estimates of a second SOC of the sensing device, and blend the sets of estimates to generate a blended SOC relation describing the second SOC of the sensing device as a function of the first SOC of the battery assembly. The estimator is also configured to filter the set of first estimates using the blended SOC relation to generate a final estimate of the SOC of the battery assembly.

Claims (138)

1 . A system for estimating a state of charge (SOC) of a battery assembly, comprising:

a sensing device connected in series to a battery cell of the battery assembly, the battery assembly having a first battery chemistry, the sensing device including a measurement cell having a second battery chemistry that is different than the first battery chemistry;

an estimator configured to perform:

acquiring a set of first estimates of a first state of charge (SOC) of the battery assembly, the set of first estimates based on a measurement of a parameter of the battery assembly performed at a set of sample times;

acquiring a set of second estimates of a second SOC of the sensing device, the set of second estimates based on a measurement of the parameter of the sensing device at the set of sample times;

blending the set of first estimates and the set of second estimates to generate a blended SOC relation, the blended SOC relation describing the second SOC of the sensing device as a function of the first SOC of the battery assembly; and

filtering the set of first estimates using the blended SOC relation to generate a final estimate of the SOC of the battery assembly, the final estimate having an accuracy that is greater than an accuracy of the set of first estimates; and

a charger module configured to control a charging operation based on the final estimate of the SOC of the battery assembly.

2 . The system of claim 1 , wherein the filtering is performed using a Kalman filter algorithm.

3 . The system of claim 1 , wherein the blending is based on a ratio of a capacity of the battery assembly to a capacity of the sensing device.

4 . The system of claim 3 , wherein the blended SOC relation is represented by:

SOC

SD

(

k

)

=

(

CAP

BP

/

CAP

SD

)

*

SOC

BP

(

k

)

+

d

%

+

v

(

i

,

T

)

,

wherein SOC SD (k) is a state of charge of the sensing device for a plurality of measurement times k, CAP BP is a capacity of the battery assembly, CAP SD is a capacity of the sensing device, d % is a minimum charge offset, SOC BP (k) is a state of the charge of the battery assembly, and v(i,T) is a noise model of the sensing device based on a current i and a temperature T.

5 . The system of claim 1 , wherein the set of first estimates is acquired based on a Coulomb counting process.

6 . The system of claim 5 , wherein the set of second estimates is acquired based on a Coulomb counting process in combination with a battery state estimation process.

7 . The system of claim 1 , wherein the estimator is further configured to determine a state of health of the battery assembly based on a capacity of the sensing device, a first offset value and a second offset value, the first offset value based on a difference between a SOC of the sensing device in a rest condition and a SOC of the battery cell in the rest condition, the second offset value based on a difference between a maximum SOC of the sensing device and a measured SOC of the sensing device.

8 . The system of claim 1 , wherein the battery cell is part of a battery pack of a vehicle.

9 . A method of estimating a state of charge (SOC) of a battery assembly, comprising:

acquiring a set of first estimates of a first state of charge (SOC) of the battery assembly, the battery assembly having a first battery chemistry, the set of first estimates based on a measurement of a parameter of the battery assembly performed at a set of sample times;

acquiring a set of second estimates of a second SOC of a sensing device connected in series to a battery cell of the battery assembly, the sensing device including a measurement cell having a second battery chemistry that is different than the first battery chemistry, the set of second estimates based on a measurement of the parameter of the sensing device at the set of sample times;

blending the set of first estimates and the set of second estimates to generate a blended SOC relation, the blended SOC relation describing the second SOC of the sensing device as a function of the first SOC of the battery assembly;

filtering the set of first estimates using the blended SOC relation to generate a final estimate of the SOC of the battery assembly, the final estimate having an accuracy that is greater than an accuracy of the set of first estimates; and

controlling a charging operation based on the final estimate of the SOC of the battery assembly.

10 . The method of claim 9 , wherein the filtering is performed using a Kalman filter algorithm.

11 . The method of claim 9 , wherein the blending is based on a ratio of a capacity of the battery assembly to a capacity of the sensing device.

12 . The method of claim 11 , wherein the blended SOC relation is represented by:

SOC

SD

(

k

)

=

(

CAP

BP

/

CAP

SD

)

*

SOC

BP

(

k

)

+

d

%

+

v

(

i

,

T

)

,

wherein SOC SD (k) is a state of charge of the sensing device for a plurality of measurement times k, CAP BP is a capacity of the battery assembly, CAP SD is a capacity of the sensing device, d % is a minimum charge offset, SOC BP (k) is a state of the charge of the battery assembly, and v(i,T) is a noise model of the sensing device based on a current i and a temperature T.

13 . The method of claim 9 , wherein the set of first estimates is acquired based on a Coulomb counting process.

14 . The method of claim 13 , wherein the set of second estimates is acquired based on a Coulomb counting process in combination with a battery state estimation process.

15 . The method of claim 9 , further comprising determining a state of health of the battery assembly based on a capacity of the sensing device, a first offset value and a second offset value, the first offset value based on a difference between a SOC of the sensing device in a rest condition and a SOC of the battery cell in the rest condition, the second offset value based on a difference between a maximum SOC of the sensing device and a measured SOC of the sensing device.

16 . The method of claim 9 , wherein the battery cell is part of a battery pack of a vehicle.

17 . A vehicle system comprising:

a memory having computer readable instructions; and

a processing device for executing the computer readable instructions, the computer readable instructions controlling the processing device to perform a method including:

acquiring a set of first estimates of a first state of charge (SOC) of a battery assembly, the battery assembly having a first battery chemistry, the set of first estimates based on a measurement of a parameter of the battery assembly performed at a set of sample times;

acquiring a set of second estimates of a second SOC of a sensing device connected in series to a battery cell of the battery assembly, the sensing device including a measurement cell having a second battery chemistry that is different than the first battery chemistry, the set of second estimates based on a measurement of the parameter of the sensing device at the set of sample times;

blending the set of first estimates and the set of second estimates to generate a blended SOC relation, the blended SOC relation describing the second SOC of the sensing device as a function of the first SOC of the battery assembly;

filtering the set of first estimates using the blended SOC relation to generate a final estimate of the SOC of the battery assembly, the final estimate having an accuracy that is greater than an accuracy of the set of first estimates; and

controlling a charging operation based on the final estimate of the SOC of the battery assembly.

18 . The vehicle system of claim 17 , wherein the filtering is performed using a Kalman filter algorithm.

19 . The vehicle system of claim 17 , wherein the blending is based on a ratio of a capacity of the battery assembly to a capacity of the sensing device, and the blended SOC relation is represented by:

SOC

SD

(

k

)

=

(

CAP

BP

/

CAP

SD

)

*

SOC

BP

(

k

)

+

d

%

+

v

(

i

,

T

)

,

wherein SOC SD (k) is a state of charge of the sensing device for a plurality of measurement times k, CAP BP is a capacity of the battery assembly, CAP SD is a capacity of the sensing device, d % is a minimum charge offset, SOC BP (k) is a state of the charge of the battery assembly, and v(i,T) is a noise model of the sensing device based on a current i and a temperature T.

20 . The vehicle system of claim 17 , wherein the set of first estimates is acquired based on a Coulomb counting process, and the set of second estimates is acquired based on a Coulomb counting process in combination with a battery state estimation process.