Systems and methods for machine learning enabled fault detection in rechargeable batteries

1. A computer-implemented method comprising:

receiving, from one or more sensors associated with a battery pack, one or more measurements pertaining to voltage, temperature, or both;

transforming the one or more measurements into a time-series sequential window flow format and compressing the time-series sequential window flow format of the one or more measurements from a higher dimension level to a lower dimension level comprising time-variant change details of the one or more measurements extracted using a sliding window associated with the time-series sequential window flow format;

determining, based on the one or more measurements, a voltage score and a temperature score;

predicting, based on the voltage score and the temperature score, whether the battery pack is experiencing a fault condition comprising a short circuit, a thermal condition, or both, wherein the prediction is performed by an artificial intelligence engine; and

responsive to predicting the battery pack is experiencing the fault condition, controlling, via a processing device, an operating parameter of the battery pack to cause the battery pack to consume a different amount of energy.

2. The computer-implemented method of claim 1 , wherein the one or more measurements are received while the battery pack is charging.

3. The computer-implemented method of claim 1 , wherein the battery pack comprises a plurality of cells and the one or more measurements are received for each of the plurality of cells, and the method further comprises comparing the one or more measurements for each of the plurality of cells.

4. The computer-implemented method of claim 1 , further comprising transmitting the one or more measurements to a cloud-based computing system, wherein the cloud- based computing system uses the one or more measurements to train one or more cloud-based machine learning models configured to predict the fault condition.

5. The computer-implemented method of claim 1 , wherein the voltage score and the temperature score are z-score values that represent cell-to-cell variation in the battery pack.

6. The computer-implemented method of claim 1 , responsive to predicting the battery pack is experiencing the fault condition, further comprising scheduling maintenance of the battery pack, repairing the battery pack, replacing the battery pack, or some combination thereof.

7. One or more non-transitory, tangible computer-readable media storing instructions that, when executed, cause one or more processing devices to:

receive, from one or more sensors associated with a battery pack, one or more measurements pertaining to voltage, temperature, or both;

transform the one or more measurements into a time-series sequential window flow format and compressing the time-series sequential window flow format of the one or more measurements from a higher dimension level to a lower dimension level comprising time-variant change details of the one or more measurements extracted using a sliding window associated with the time-series sequential window flow format;

determine, based on the one or more measurements, a voltage score and a temperature score;

predict, based on the voltage score and the temperature score, whether the battery pack is experiencing a fault condition comprising a short circuit, a thermal condition, or both, wherein the prediction is performed by an artificial intelligence engine; and

responsive to predicting the battery pack is experiencing the fault condition, controlling, via a processing device, an operating parameter of the battery pack to cause the battery pack to consume a different amount of energy.

8. The one or more computer-readable media of claim 7 , wherein the one or more measurements are received while the battery pack is charging.

9. The one or more computer-readable media of claim 7 , wherein the battery pack comprises a plurality of cells and the one or more measurements are received for each of the plurality of cells, and the one or more processing devices are further configured to compare the one or more measurements for each of the plurality of cells.

10. The one or more computer-readable media of claim 7 , further comprising transmitting the one or more measurements to a cloud-based computing system, wherein the cloud- based computing system uses the one or more measurements to train one or more cloud-based machine learning models configured to predict the fault condition.

11. The one or more computer-readable media of claim 7 , wherein the voltage score and the temperature score are z-score values that represent cell-to-cell variation in the battery pack.

12. The one or more computer-readable media of claim 7 , responsive to predicting the battery pack is experiencing the fault condition, scheduling maintenance of the battery pack, repairing the battery pack, replacing the battery pack, or some combination thereof.

13. A system comprising:

one or more memory devices storing instructions; and

one or more processing devices communicatively coupled to the one or more memory devices, wherein the one or more processing devices execute the instructions to:

receive, from one or more sensors associated with a battery pack, one or more measurements pertaining to voltage, temperature, or both;

transform the one or more measurements into a time-series sequential window flow format and compressing the time-series sequential window flow format of the one or more measurements from a higher dimension level to a lower dimension level comprising time-variant change details of the one or more measurements extracted using a sliding window associated with the time-series sequential window flow format;

determine, based on the one or more measurements, a voltage score and a temperature score;

predict, based on the voltage score and the temperature score, whether the battery pack is experiencing a fault condition comprising a short circuit, a thermal condition, or both, wherein the prediction is performed by one or more trained machine learning models; and

responsive to predicting the battery pack is experiencing the fault condition, controlling, via a processing device, an operating parameter of the battery pack to cause the battery pack to consume a different amount of energy.

14. The system of claim 13 , wherein the one or more measurements are received while the battery pack is charging.

15. The system of claim 13 , wherein the battery pack comprises a plurality of cells and the one or more measurements are received for each of the plurality of cells, and the one or more processing devices are further configured to compare the one or more measurements for each of the plurality of cells.

16. The system of claim 13 , further comprising transmitting the one or more measurements to a cloud-based computing system, wherein the cloud-based computing system uses the one or more measurements to train one or more cloud-based machine learning models configured to predict the fault condition.

17. The system of claim 13 , wherein the voltage score and the temperature score are z-score values that represent cell-to-cell variation in the battery pack.