Techniques for robust battery state estimation

Patent metadata

Field	Value
Publication number	US-10408880-B2
Application number	US-201414463016-A
Country	US
Kind code	B2
Filing date	Aug 19, 2014
Priority date	Aug 19, 2014
Publication date	Sep 10, 2019
Grant date	Sep 10, 2019

Abstract

More accurate and robust battery state estimation (BSE) techniques for a battery system of an electrified vehicle include estimating a current bias or offset generated by a current sensor and then adjusting the measured current to compensate for the estimated current bias. The techniques obtain nominal parameters for a battery model of the battery system based on a measured temperature and an estimated open circuit voltage (OCV). The techniques use these nominal parameters and the corrected measured current to estimate the OCV, a capacity, and an impedance of the battery system. The techniques utilize the OCV to estimate a state of charge (SOC) of the battery system. The techniques also estimate a state of health (SOH) of the battery system based on its estimated capacity and impedance. The techniques then control the electrified vehicle based on the SOC and/or the SOH.

First claim

What is claimed is:

1. A method, comprising:
receiving, at a controller of an electrified vehicle, measurements of current, voltage, and temperature of a battery system of the electrified vehicle, the measured current being obtained by a current sensor;
obtaining, at the controller, a set of parameters for an equivalent circuit model of the battery system based on the measured current, voltage, and temperature, the set of parameters including at least one of resistance, capacitance, and capacity;
detecting, at the controller, a low current period during which the measured current is less than a predetermined threshold and having a duration that is sufficiently long to provide a stable low current such that a current bias generated by the current sensor is assumed to be constant;
in response to detecting the low current period:
converting, at the controller, the equivalent circuit model from a continuous-time domain to a discrete-time domain to obtain state-space matrices for the measured current, and
applying, at the controller, a Kalman filter (KF) algorithm to the state-space matrices to estimate the current bias generated by the current sensor based on the measured voltage and the set of parameters;

adjusting, at the controller, the measured current based on the estimated current bias to obtain a corrected measured current;
estimating, at the controller, an open circuit voltage (OCV) of the battery system based on the measured voltage, the set of parameters, and the corrected measured current;
estimating, at the controller, a state of charge (SOC) of the battery system based on the estimated OCV of the battery system; and
controlling, by the controller, at least one of the battery system and an electric motor of the electrified vehicle based on the estimated SOC, thereby compensating for the estimated current bias generated by the current sensor and improving usage of the battery system.

2. The method of claim 1 , further comprising, based on the measured voltage, the set of parameters, and the corrected measured current, estimating, at the controller, at least one of (i) a capacity of the battery system and (ii) an impedance of the battery system.

3. The method of claim 2 , further comprising estimating, at the controller, a state of health (SOH) of the battery system based on its capacity and its impedance.

4. The method of claim 3 , further comprising controlling, by the controller, at least one of the battery system and an electric motor of the electrified vehicle based on the estimated SOH of the battery system.

5. The method of claim 2 , further comprising, as a function of the estimated SOC, performing, at the controller, recursive (i) obtaining of the set of parameters and (ii) estimation of the OCV, the capacity, and the impedance.

6. The method of claim 5 , wherein estimating the SOC based on the estimated OCV includes utilizing an OCV-SOC curve model, and wherein the function is a ratio of (i) change in OCV to (ii) change in SOC.

7. The method of claim 2 , wherein estimating each of the OCV, the capacity, and the impedance based on the measured voltage, the set of parameters, and the corrected measured current further includes:
performing, at the controller, an LDL factorization to increase the accuracy of each respective estimation algorithm; and
utilizing, at the controller, each respective LDL factorized estimation algorithm with the measured voltage, the set of parameters, and the corrected measured current to obtain the estimated OCV, the estimated capacity, and the estimated impedance.

8. The method of claim 7 , wherein each respective estimation algorithm is one of (i) a recursive least squares (RLS) adaptive filter algorithm, (ii) a Kalman filter (KF) algorithm, and (iii) an extended KF (EKF) algorithm.

9. The method of claim 1 , further comprising applying, at the controller, predetermined constraints or ranges to the values of the set of parameters for the battery system.

10. The method of claim 1 , wherein at least some of the set of parameters are different for charging and discharging of the battery system.

11. An electrified vehicle, comprising:
an electric motor configured to propel the electrified vehicle;
a battery system configured to power the electric motor;
a current sensor configured to measure a current of the battery system; and
a controller configured to:
receive the measured current from the current sensor;
receive a measured voltage of the battery system and a measured temperature of the battery system;
obtain a set of parameters for an equivalent circuit model of the battery system based on the measured current, voltage, and temperature, the set of parameters including at least one of resistance, capacitance, and capacity;
detect a low current period during which the measured current is less than a predetermined threshold and having a duration that is sufficiently long to provide a stable low current such that a current bias generated by the current sensor is assumed to be constant;
in response to detecting the low current period:
convert the equivalent circuit model from a continuous-time domain to a discrete-time domain to obtain state-space matrices for the measured current, and
apply a Kalman filter (KF) algorithm to the state-space matrices to estimate the current bias generated by the current sensor based on the measured voltage and the set of parameters;

adjust the measured current based on the estimated bias or offset of the current sensor to obtain a corrected measured current;
estimate an open circuit voltage (OCV) of the battery system based on the measured voltage, the set of parameters, and the corrected measured current;
estimate a state of charge (SOC) of the battery system based on the estimated OCV of the battery system; and
control at least one of the battery system and the electric motor based on the estimated SOC, thereby compensating for the estimated current bias generated by the current sensor and improving usage of the battery system.

12. The electrified vehicle of claim 11 , wherein based on the measured voltage, the set of parameters, and the corrected measured current, the controller is further configured to estimate at least one of (i) a capacity of the battery system and (ii) an impedance of the battery system.

13. The electrified vehicle of claim 12 , wherein the controller is further configured to estimate a state of health (SOH) of the battery system based on its capacity and its impedance.

14. The electrified vehicle of claim 13 , wherein the controller is further configured to control at least one of the battery system and the electric motor based on at least one of the estimated SOC and the estimated SOH of the battery system.

Assignees

• Lin Jian
• Yang Hong
• Malysz Pawel
• Ye Jin
• Gu Ran
• Emadi Ali
• Fca Us Llc

Inventors

• Lin Jian
• Yang Hong
• Malysz Pawel
• Ye Jin
• Gu Ran
• Emadi Ali

Classifications

• G01R31/3842

  combining voltage and current measurements · CPC title

• B60L58/10

  for monitoring or controlling batteries · CPC title

• G01R31/367 Primary

  Software therefor, e.g. for battery testing using modelling or look-up tables · CPC title

• G01R31/392

  Determining battery ageing or deterioration, e.g. state of health · CPC title

• Y02T10/7005

  Cross-Sectional Technologies · mapped topic