Smart battery management systems

Patent metadata

Field	Value
Publication number	US-12107240-B2
Application number	US-202318198268-A
Country	US
Kind code	B2
Filing date	May 16, 2023
Priority date	Feb 28, 2019
Publication date	Oct 1, 2024
Grant date	Oct 1, 2024

Abstract

Various implementations of a smart battery management system are provided. An example method includes identifying sensor data of a cell in a battery system; predicting, based on the sensor data, a failure event of the cell; and preventing the failure event by activating a control circuit connected to the cell.

First claim

The invention claimed is:

1. A method for managing a battery system, comprising:
receiving sensor data from sensors configured to measure one or more metrics of a cell in a battery system;
predicting a failure event of the cell; and
preventing the failure event by activating a control circuit coupled to the cell, wherein the sensor data sensors configured to measure one or more metrics of a plurality of cells in a battery,
wherein the metrics comprise at least movement of the cell obtained from at least one movement sensor, and
wherein the metrics further comprise at least one of a pressure obtained from at least one pressure sensor, a temperature obtained from at least one temperature sensor, a voltage obtained from at least one voltage sensor, a current obtained from at least one current sensor, a resistance obtained from at least one resistance sensor, an impedance obtained from at least one impedance sensor, or a capacitance obtained from at least one capacitance sensor, each associated with the cell.

2. The method of claim 1 , wherein the step of predicting is based on the sensor data, the failure event is predicted by a neural network.

3. The method of claim 2 , the sensor data being first sensor data, the cell being a first cell, and the failure event being a first failure event, the method further comprising:
identifying second sensor data of a second cell, the second sensor data being obtained during at least one second failure event of the second cell; and
training the neural network based on the second sensor data and an indication of the at least one second failure event.

4. The method of claim 1 , wherein the failure event comprises thermal runaway of the cell.

5. The method of claim 1 , wherein activating the control circuit comprises:
causing the control circuit to modify a current flowing through the cell, including causing to disconnect the cell from one or more additional cells in the battery system.

6. The method of claim 1 , wherein preventing the failure event comprises:
providing the acceleration data obtained from the at least one movement sensor.

7. The method of claim 6 , further comprising:
predicting by inputting data from the at least one movement sensor into the trained neural network, a failure event of the cell based on the associated movement data from the cell.

8. The method of claim 7 , further comprising:
disconnecting the cell from the one or more additional cells in response to the prediction of the trained neural network.

9. A Smart Battery Management System, comprising:
at least one processor; and
non-transient memory storing instructions that, when executed by the at least one processor cause the at least one processor to perform operations comprising:
identifying sensor data of a cell in a battery system;
predicting, based on the sensor data, a failure event of the cell; and
preventing the failure event by activating a control circuit connected to the cell,
wherein the metrics comprise at least movement of the cell obtained from at least one movement sensor, and
wherein the metrics further comprise at least one of a pressure obtained from at least one pressure sensor, a temperature obtained from at least one temperature sensor, a voltage obtained from at least one voltage sensor, a current obtained from at least one current sensor, a resistance obtained from at least one resistance sensor, an impedance obtained from at least one impedance sensor, or a capacitance obtained from at least one capacitance sensor, each associated with the cell.

10. The system of claim 9 , wherein predicting is based on the sensor data, the failure event is predicted by a neural network.

11. The system of claim 10 , the sensor data being first sensor data, the cell being a first cell, and the failure event being a first failure event, the processor further performs:
identifying second sensor data of a second cell, the second sensor data being obtained during at least one second failure event of the second cell; and
training the neural network based on the second sensor data and an indication of the at least one second failure event.

12. The system of claim 9 , wherein the failure event comprises thermal runaway of the cell.

13. The system of claim 9 , wherein activating the control circuit comprises:
causing the control circuit to modify a current flowing through the cell, including causing to disconnect the cell from one or more additional cells in the battery system.

14. The system of claim 9 , wherein preventing the failure event comprises:
providing acceleration data obtained from the at least one movement sensor.

15. The system of claim 14 , further comprising:
by inputting data from the at least one movement sensor into a trained neural network, thereby predicting the failure event of the cell based on the associated movement data from the cell.

16. The system of claim 15 , further comprising:
disconnecting the cell from the one or more additional cells in response to the prediction of the trained neural network.

17. A method for managing a battery system, comprising:
receiving sensor data from sensors configured to measure one or more metrics of a cell in a battery system;
predicting a failure event of the cell; and
preventing the failure event by activating a control circuit coupled to the cell, wherein the sensor data sensors configured to measure one or more metrics of a plurality of cells in a battery,
wherein the step of predicting a failure event is based on the sensor data and predicted by a neural network, and
wherein the sensor data being first sensor data, the cell being a first cell, and the failure event being a first failure event, the method further comprising:
identifying second sensor data of a second cell, the second sensor data being obtained during at least one second failure event of the second cell; and
training the neural network based on the second sensor data and an indication of the at least one second failure event.

18. A Smart Battery Management System, comprising:
at least one processor; and
non-transient memory storing instructions that, when executed by the at least one processor cause the at least one processor to perform operations comprising:
identifying sensor data of a cell in a battery system;
predicting, based on the sensor data, a failure event of the cell; and

preventing the failure event by activating a control circuit connected to the cell,
wherein predicting is based on the sensor data, the failure event is predicted by a neural network, and
wherein the sensor data being first sensor data, the cell being a first cell, and the failure event being a first failure event, the processor further performs:
identifying second sensor data of a second cell, the second sensor data being obtained during at least one second failure event of the second cell; and
training the neural network based on the second sensor data and an indication of the at least one second failure event.

19. A Smart Battery Management System, comprising:
at least one processor; and
non-transient memory storing instructions that, when executed by the at least one processor cause the at least one processor to perform operations comprising:
identifying sensor data of a cell in a battery system;
predicting, based on the sensor data, a failure event of the cell; and
preventing the failure event by activating a control circuit connected to the cell,
wherein preventing the failure event comprises:
providing the acceleration data obtained fr

Assignees

• Purdue Research Foundation

Inventors

• Tomar Vikas
• Adams Thomas Edward
• Alvarado Jonathan E
• Dietz James Eric
• Li Bing
• Neal Christian T

Classifications

• H01M50/574

  Devices or arrangements for the interruption of current · CPC title

• H01M10/486

  for measuring temperature · CPC title

• H01M10/482

  for several batteries or cells simultaneously or sequentially · CPC title

• B60L53/63

  in response to network capacity · CPC title

• H01M2010/4278

  Systems for data transfer from batteries, e.g. transfer of battery parameters to a controller, data transferred between battery controller and main controller · CPC title