System and method for controlling a battery management system

Patent metadata

Field	Value
Publication number	US-11529887-B2
Application number	US-202016751871-A
Country	US
Kind code	B2
Filing date	Jan 24, 2020
Priority date	Jan 24, 2020
Publication date	Dec 20, 2022
Grant date	Dec 20, 2022

Abstract

A battery management system includes one or more processors, a battery comprising a plurality of cells, an output device, an input device, and a memory having an input module, a battery characteristic prediction module, and an output module. The input module includes instructions that cause the one or more processors to receive a mode selection from a user via the input device. The battery characteristic prediction module includes instructions that cause the one or more processors to predict a characteristic of the battery based on the mode selection using an active machine learning model to predict the characteristic of the battery. The output module includes instructions that cause the one or more processors to output an estimated cost to the output device based on the characteristic of the battery determined by the active machine learning model.

First claim

What is claimed is:

1. A battery management system comprising:
one or more processors;
a battery comprising a plurality of cells, the battery being in communication with the one or more processors;
an output device in communication with the one or more processors;
an input device in communication with the one or more processors;
a memory in communication with the one or more processors the memory having an input module, a battery characteristic prediction module, and an output module;
wherein the input module includes instructions that, when executed by the one or more processors, cause the one or more processors to receive a mode selection from a user via the input device, the mode selection includes a selection between extending a driving range of the vehicle and increasing cycle life of the battery;
wherein the battery characteristic prediction module includes instructions that, when executed by the one or more processors, cause the one or more processors to predict a characteristic of the battery based on the mode selection, wherein the battery characteristic prediction module utilizes an active machine learning model to predict the characteristic of the battery, wherein the characteristic of the battery is a cycle life of the battery; and
wherein the output module includes instructions that, when executed by the one or more processors, cause the one or more processors to output an estimated cost to the output device based on the characteristic of the battery determined by the active machine learning model, the estimated cost being a prediction how the selection of mode selection impacts the driving range of the vehicle and the cycle life of the battery.

2. The battery management system of claim 1 , wherein the cycle life of the battery is a number of cycles until 80% of a nominal capacity of the battery.

3. The battery management system of claim 1 , further comprising:
a network access device in communication with the one or more processors:
wherein the memory further comprises a communications module having instructions that, when executed by the one or more processors, cause the one or more processors to receive updated model weights from an external system via the network access device; and
wherein the memory further comprises an active learning module having instructions that, when executed by the one or more processors, cause the one or more processors to update the active machine learning model with model weights obtained by training the active machine learning model on the external system.

4. The battery management system of claim 1 , wherein the battery management system is mounted within a vehicle.

5. The battery management system of claim 4 , wherein the battery characteristic prediction module includes instructions that, when executed by the one or more processors, cause the one or more processors to predict the characteristic of the battery based on the mode selection and a driving style of the user, the driving style of the user indicating one or more driving characteristics of the user when operating the vehicle.

6. The battery management system of claim 5 , wherein the one or more driving characteristics includes a historical distance the vehicle travels between destinations, a historical speed of the vehicle, and a charging history of the battery of the vehicle.

7. A method for managing a battery management system comprising the steps of:
receiving, by one or more processors, a mode selection from a user via an input device, the mode selection includes a selection between extending a driving range of the vehicle and increasing cycle life of the battery;
predicting, by the one or more processors, a characteristic of a battery based on the mode selection by utilizing an active machine learning model, the battery comprising a plurality of cells, wherein the characteristic of the battery is a cycle life of the battery; and
outputting, by the one or more processors, an estimated cost to an output device based on the characteristic of the battery determined by the active machine learning model, the estimated cost being a prediction how the selection of mode selection impacts the driving range of the vehicle and the cycle life of the battery.

8. The method for managing the battery management system of claim 7 , wherein the cycle life of the battery is a number of cycles until 80% of a nominal capacity of the battery.

9. The method for managing the battery management system of claim 7 , further comprising the steps of
receiving, by the one or more processors via a network access device, updated model weights from an external system via the network access device; and
updating, by the one or more processors, the active machine learning model with model weights obtained by training the active machine learning model on the external system.

10. The method for managing the battery management system of claim 7 , wherein the battery management system is mounted within a vehicle.

11. The method for managing the battery management system of claim 10 , further comprising the step of predicting the characteristic of the battery based on the mode selection and a driving style of the user, the driving style of the user indicating one or more driving characteristics of the user when operating the vehicle.

12. The method for managing the battery management system of claim 11 , wherein the one or more driving characteristics includes a historical distance the vehicle travels between destinations, a historical speed of the vehicle, and a charging history of the battery of the vehicle.

13. A non-transitory computer-readable medium for controlling a battery management system, the non-transitory computer-readable medium comprising instructions that when executed by one or more processors cause the one or more processors to:
receive a mode selection from a user via an input device, the mode selection includes a selection between extending a driving range of the vehicle and increasing cycle life of the battery;
predict a characteristic of a battery based on the mode selection by utilizing an active machine learning model, the battery comprising a plurality of cells, wherein the characteristic of the battery is a cycle life of the battery; and
output an estimated cost to an output device based on the characteristic of the battery determined by the active machine learning model, the estimated cost being a prediction how the selection of mode selection impacts the driving range of the vehicle and the cycle life of the battery.

14. The non-transitory computer-readable medium of claim 13 , wherein the cycle life of the battery is a number of cycles until 80% of a nominal capacity of the battery.

15. The non-transitory computer-readable medium of claim 13 , further comprising instructions that when executed by the one or more processors cause the one or more processors to:
receive, via a network access device, updated model weights from an external system via the network access device; and
update the active machine learning model with model weights obtained by training the active machine learning model on the external system.

Assignees

• Toyota Res Inst Inc

Inventors

• Herring Patrick K
• Balaji Gopal Chirranjeevi
• Anapolsky Abraham S

Classifications

• H01M10/482

  for several batteries or cells simultaneously or sequentially · CPC title

• B60L58/16 Primary

  responding to battery ageing, e.g. to the number of charging cycles or the state of health [SoH] · CPC title

• B60L58/18

  of two or more battery modules · CPC title

• B60L2260/50

  by future state prediction · CPC title

• G06N3/08

  Learning methods · CPC title