Battery capacity prediction system using charge and discharge cycles of a battery to predict capacity variations, and associated method

Patent metadata

Field	Value
Publication number	US-11293987-B2
Application number	US-202016822469-A
Country	US
Kind code	B2
Filing date	Mar 18, 2020
Priority date	Jun 5, 2019
Publication date	Apr 5, 2022
Grant date	Apr 5, 2022

Abstract

A system and method of predicting variations in a capacity of a battery, the system including an ADFM management device including an experimental data collector to collect at least one first piece of data about the capacity of the battery, an ADF optimizer to optimize a first calculation equation, and a virtual data generator to generate at least one second piece of data; and a server including a training unit to train an artificial intelligence model for outputting the relative capacity variation value by using the at least one first piece of data and the at least one second piece of data as training data, and a prediction unit to obtain a relative capacity variation prediction value, which is output from the artificial intelligence model when the number of charge and discharge cycles and the charge and discharge conditions of the battery are input to the artificial intelligence model.

First claim

What is claimed is:

1. A battery capacity prediction system, comprising:
an aging density function model (ADFM) management device;
and a server,
wherein the ADFM management device includes:
an experimental data collector configured to collect at least one first piece of data about the capacity of the battery according to a number of charge and discharge cycles and charge and discharge cycle conditions of the battery,
an ADF optimizer configured to optimize a first calculation equation for predicting a relative capacity variation value of the battery based on the at least one first piece of data, wherein the first calculation equation calculates the relative capacity variation value by integrating an aging density of the battery with respect to time based on the at least one first piece of data, and
a virtual data generator configured to generate at least one second piece of data about the relative capacity variation value corresponding to the number of charge and discharge cycles and the charge and discharge conditions of the battery, based on the first calculation equation;

wherein the server includes:
a training unit configured to train an artificial intelligence model for outputting the relative capacity variation value by using the at least one first piece of data and the at least one second piece of data as training data, and
a prediction unit configured to obtain a relative capacity variation prediction value, which is output from the artificial intelligence model when the number of charge and discharge cycles and the charge and discharge conditions of the battery are input to the artificial intelligence model,

collecting the at least one first piece of data includes:
performing a reference performance test (RPT) on the battery every preset number of charge and discharge cycles; and
obtaining an open circuit voltage-state of charge lookup table (OCV-SOC LUT) based on results of the RPT,

the first calculation equation is also for calculating the relative capacity variation value by calculating combinations with repetition based on a charge rate, a discharge rate, a maximum state of charge (SOC) per cycle, a minimum SOC per cycle, and a temperature of the battery, and
the optimizing of the first calculation equation includes determining the maximum SOC per cycle and the minimum SOC per cycle of the battery based on the OCV-SOC LUT.

2. A computer-implemented method of predicting variations in a capacity of a battery according to charge and discharge cycles of the battery, the method comprising:
collecting at least one first piece of data about the capacity of the battery according to a number of charge and discharge cycles and charge and discharge cycle conditions of the battery;
optimizing a first calculation equation for predicting a relative capacity variation value of the battery based on the at least one first piece of data, wherein the first calculation equation calculates the relative capacity variation value by integrating an aging density of the battery with respect to time based on the at least one first piece of data;
generating at least one second piece of data about the relative capacity variation value corresponding to the number of charge and discharge cycles and the charge and discharge conditions of the battery, based on the first calculation equation;
training an artificial intelligence model for outputting the relative capacity variation value by using the at least one first piece of data and the at least one second piece of data as training data; and
obtaining a relative capacity variation prediction value, which is output from the artificial intelligence model when the number of charge and discharge cycles and the charge and discharge conditions of the battery are input to the artificial intelligence model,
wherein:
collecting the at least one first piece of data includes:
performing a reference performance test (RPT) on the battery every preset number of charge and discharge cycles; and
obtaining an open circuit voltage-state of charge lookup table (OCV-SOC LUT) based on results of the RPT,

the first calculation equation is for calculating the relative capacity variation value by calculating combinations with repetition based on a charge rate, a discharge rate, a maximum state of charge (SOC) per cycle, a minimum SOC per cycle, and a temperature of the battery, and
the optimizing of the first calculation equation includes determining the maximum SOC per cycle and the minimum SOC per cycle of the battery based on the OCV-SOC LUT.

3. The computer-implemented method as claimed in claim 2 , wherein:
the charge and discharge conditions of the battery include a constant current charge period, a constant voltage charge period, a first rest period, a constant current discharge period, and a second rest period, and
collecting the at least one first piece of data further includes collecting information about the capacity of the battery according to the charge and discharge conditions.

4. The computer-implemented method as claimed in claim 2 , wherein optimizing the first calculation equation further includes:
obtaining a parameter of the first calculation equation, the parameter minimizing an error between a first relative capacity loss amount obtained based on the first calculation equation and a second relative capacity loss amount obtained based on the at least one first piece of data; and
updating the first calculation equation based on the parameter.

Assignees

• Samsung Sdi Co Ltd
• Postech Res & Business Dev Found

Inventors

• Lee Gyeonghwan
• Kim Minho
• Han Soohee

Classifications

• Y02E60/10

  Energy storage using batteries · CPC title

• G01R31/392

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

• G01R31/367 Primary

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

• G01R31/382

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

• G01R31/3648 Primary

  comprising digital calculation means, e.g. for performing an algorithm · CPC title