Method and apparatus for estimating state of battery

Patent metadata

Field	Value
Publication number	US-2016231386-A1
Application number	US-201615010977-A
Country	US
Kind code	A1
Filing date	Jan 29, 2016
Priority date	Feb 6, 2015
Publication date	Aug 11, 2016
Grant date	—

Abstract

A battery state estimating apparatus includes a learner configured to learn a battery state estimation model comprising a label corresponding to a battery state; and a state estimator configured to estimate a battery state of a target battery using the battery state estimation model.

First claim

What is claimed is:

1 . A battery state estimating apparatus comprising:
a learner configured to learn a battery state estimation model comprising a label corresponding to a battery state; and a state estimator configured to estimate a battery state of a target battery using the battery state estimation model.

2 . The apparatus of claim 1 , wherein the learner comprises:
a segment data generator configured to generate segment data by segmenting a battery sensor signal of a reference battery based on a predetermined time interval; a labeler configured to assign, to each of the segment data, a respective label corresponding to the battery state; and a pattern extractor configured to extract a feature pattern of each label from the labeled segment data.

3 . The apparatus of claim 2 , wherein the learner further comprises a battery state estimation model generator configured to generate, as the battery state estimation model, a feature space transformation model corresponding to the feature pattern of each label using a discriminant analysis method.

4 . The apparatus of claim 2 , wherein the battery sensor signal comprises any one or any combination of any two or more of voltage data, current data, temperature data, and pressure data of the reference battery in a time-series space.

5 . The apparatus of claim 2 , wherein the segment data generator comprises a preprocessor configured to correct the battery sensor signal at a fixed sampling interval before the segmenting of the battery sensor signal.

6 . The apparatus of claim 1 , wherein the state estimator comprises:
a segment data generator configured to generate segment data by segmenting a battery sensor signal of the target battery based on a predetermined time interval; a pattern matcher configured to perform pattern matching between the battery state estimation model and the segment data of the target battery; and a battery state estimator configured to estimate the battery state of the target battery based on a result of the pattern matching.

7 . The apparatus of claim 6 , wherein the pattern matcher is further configured to perform the pattern matching between a feature pattern of a battery state estimation model corresponding to a battery state to be estimated and a data pattern of the segment data of the target battery; and
the battery state estimator is further configured to estimate, as the battery state of the target battery, a battery state corresponding to a feature pattern most closely matching the data pattern of the segment data of the target battery.

8 . The apparatus of claim 1 , wherein the battery state comprises a state of charge (SoC);
the learner is further configured to learn an SoC estimation model comprising SoC labels respectively corresponding to SoC sections based on a predetermined interval; and the state estimator is further configured to estimate an SoC of the target battery by comparing the SoC estimation model to segment data of the target battery obtained by segmenting a battery sensor signal of the target battery based on a predetermined time interval.

9 . The apparatus of claim 8 , wherein the learner is further configured to extract a feature pattern of each SoC label from the SoC estimation model; and
the state estimator is further configured to perform pattern matching between a data pattern of the segment data and the feature pattern of each SoC label, and estimate, as the SoC of the target battery, an SoC section corresponding to a feature pattern most closely matching the data pattern of the segment data.

10 . The apparatus of claim 1 , wherein the learner comprises a sensor state estimation model learner configured to learn a sensor state estimation model comprising labels respectively corresponding to sensor states; and
the state estimator comprises a sensor state estimator configured to estimate a sensor state of a sensor measuring a battery sensor signal of the target battery using the sensor state estimation model.

11 . The apparatus of claim 10 , wherein the learner further comprises a sensor state feature pattern extractor configured to extract a feature pattern of each of the labels corresponding to the sensor states; and
the sensor state estimator is further configured to perform pattern matching between the feature pattern of each of the labels corresponding to the sensor states and a data pattern of segment data of the target battery obtained by segmenting the battery sensor signal of the target battery based on a predetermined time interval, and estimate a sensor state corresponding to the feature pattern as the sensor state of the sensor measuring the battery sensor signal of the target battery based on a result of the pattern matching.

12 . A battery state estimating method comprising:
learning a battery state estimation model comprising a label corresponding to a battery state; and estimating a battery state of a target battery using the battery state estimation model.

13 . The method of claim 12 , wherein the learning of the battery state estimation model comprises:
generating segment data by segmenting a battery sensor signal of a reference battery based on a predetermined time interval; assigning, to each of the segment data, a label corresponding to the battery state; and extracting a feature pattern of each label from the labeled segment data.

14 . The method of claim 13 , wherein the learning of the battery state estimation model further comprises generating, as the battery state estimation model, a feature space transformation model corresponding to the feature pattern using a discriminant analysis method.

15 . The method of claim 12 , wherein the estimating of the battery state comprises:
generating segment data by segmenting a battery sensor signal of the target battery based on a predetermined time interval; performing pattern matching between the battery state estimation model and the segment data of the target battery; and estimating the battery state of the target battery based on a result of the pattern matching.

16 . The method of claim 15 , wherein the performing of the pattern matching comprises performing pattern matching between a feature pattern of a battery state estimation model corresponding to a battery state to be estimated and a data pattern of the segment data of the target battery; and
the estimating of the battery state comprises estimating, as the battery state of the target battery, a battery state corresponding to a feature pattern most closely matching the data pattern of the segment data of the target battery.

17 . The method of claim 12 , wherein the battery state comprises a state of charge (SoC);
the learning of the battery state estimation model comprises learning an SoC estimation model comprising SoC labels respectively corresponding to SoC sections based on a predetermined interval; and the estimating of the battery state comprises estimating an SoC of the target battery by comparing the SoC estimation model to segment data of the target battery obtained by segmenting a battery sensor signal of the target battery based on a predetermined time interval.

18 . The method of claim 17 , wherein the learning of the battery state estimation model comprises extracting a feature pattern of each SoC label from the SoC estimation model;
and the estimating of the battery state further comprises:
performing pattern matching between a data pattern of the segment data and the feature pattern of each SoC label; and
estimating, as the SoC of the target batt

Assignees

• Samsung Electronics Co Ltd

Inventors

• Sung Jaemo
• Park Jeonghyun

Classifications

• H01M10/425 Primary

  Structural combination with electronic components, e.g. electronic circuits integrated to the outside of the casing (printed circuits H05K1/00) · CPC title

• G01R31/382

  Arrangements for monitoring battery or accumulator variables, e.g. SoC · CPC title

• G01R31/367 Primary

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

• H01M10/488

  Cells or batteries combined with indicating means for external visualization of the condition, e.g. by change of colour or of light density · CPC title

• H01M2010/4271

  Battery management systems including electronic circuits, e.g. control of current or voltage to keep battery in healthy state, cell balancing · CPC title