Systems and methods for battery health prediction with on-device learning

1 . A method for operating a battery management system of a mobile electronic device, comprising:

accessing, from a remote server, a general battery-health model corresponding to the mobile electronic device, wherein the general battery-health model is a first deep neural network trained to calculate a state-of-health value for a battery of the mobile electronic device based on historic battery output capacity, historic battery output voltage, and historic battery output current associated with the mobile electronic device;

accessing first real-time battery charging data from the mobile electronic device for a plurality of battery charging modes, wherein the first real-time battery charging data comprises battery input voltage and battery input current during each of the plurality of battery charging modes;

identifying a plurality of data subgroups of battery charging data based on the first real-time battery charging data, wherein each data subgroup is associated with one of the battery charging modes;

receiving power from a power source electrically coupled to the battery of the mobile electronic device and being configured to charge the battery of the mobile electronic device; and

in response to the battery of the mobile electronic device achieving a maximum charge:

training, using the general battery-health model as a starting model, a first customized battery-health model in a cascaded workflow using each of the plurality of data subgroups of battery charging data for the plurality of battery charging modes, wherein the first customized battery-health model is a second deep neural network trained to calculate state-of-health values associated with the battery of the mobile electronic device for the plurality of battery charging modes, respectively;

generating, using the first customized battery-health model, a plurality of first state-of-health values associated with the battery of the mobile electronic device for the plurality of battery charging modes, respectively, based on real-time battery output capacity, real-time battery output voltage, and real-time battery output current, wherein the generated plurality of first state-of-health values enable the battery management system to prolong a battery life of the battery of the mobile electronic device; and

alleviating a potential unexpected deactivation of the mobile electronic device based at least in part on the generated plurality of first state-of-health values.

2 . The method of claim 1 , further comprising:

determining a safety risk associated with the mobile electronic device based on the plurality of first state-of-health values associated with the plurality of battery charging modes; and

presenting, by the mobile electronic device, an indication of the determined safety risk and the plurality of first state-of-health values associated with the mobile electronic device for the plurality of battery charging modes.

3 . The method of claim 2 , further comprising:

generating a work order of maintenance based on the safety risk associated with the mobile electronic device and the plurality of first state-of-health values associated with the mobile electronic device for the plurality of battery charging modes; and

sending, to the remote server, the work order of maintenance for the mobile electronic device.

4 . The method of claim 1 , further comprising, by the remote server:

accessing the historic battery output capacity, historic battery output voltage, and historic output current associated with the mobile electronic device;

training the general battery-health model based on the historic battery output capacity, historic battery output voltage, and historic battery output current associated with the mobile electronic device; and

sending, from the remote server to the mobile electronic device, the general battery-health model via a wireless network.

5 . The method of claim 1 , further comprising:

accessing second real-time battery charging data from the mobile electronic device from a latest charging cycle; wherein the second real-time battery charging data comprises battery input voltage and battery input current during each of the plurality of battery charging modes from the latest charging cycle;

training, using the first customized battery-health model as a starting model, a second customized battery-health model using the second real-time battery charging data from the mobile electronic device from the latest charging cycle; and

determining, using the second customized battery-health model, a plurality of second state-of-health values associated with the mobile electronic device for the plurality of battery charging modes, respectively, based on the real-time battery output capacity, the real-time battery output voltage, and the real-time battery output current.

6 . The method of claim 1 ,

wherein the plurality of battery charging modes includes a fast charging mode and a slow charging mode, and

wherein the general battery-health model and customized battery-health model are based on a lithium-ion battery associated with the mobile electronic device.

7 . The method of claim 6 ,

wherein the real-time battery charging data has a first sampling rate of the fast charging mode and a second sampling rate of the slow charging mode, wherein the first sampling rate is at a higher frequency than the second sampling rate, and

wherein the method further comprises:

applying over-sampling to the real-time battery charging data for the slow charging mode to match a sampling rate of the battery charging data for the fast charging mode.

8 . The method of claim 6 ,

wherein the real-time battery charging data has a first sampling rate of the slow charging mode and a second sampling rate of the fast charging mode, wherein the first sampling rate is at a higher frequency than the second sampling rate, and

wherein the method further comprises:

applying over-sampling to the real-time battery charging data for the fast charging mode to match a sampling rate of the battery charging data for the slow charging mode.

9 . A mobile electronic device, comprising:

one or more displays;

one or more non-transitory computer-readable storage media including instructions; and

one or more processors coupled to the storage media, the one or more processors configured to execute the instructions to:

access, from a remote server, a general battery-health model corresponding to the mobile electronic device, wherein the general battery-health model is a first deep neural network trained to calculate a state-of-health value for a battery of the mobile electronic device based on historic battery output capacity, historic battery output voltage, and historic battery output current associated with the mobile electronic device;

access first real-time battery charging data from the mobile electronic device for a plurality of battery charging modes, wherein the first real-time battery charging data comprises battery input voltage and battery input current during each of the plurality of battery charging modes;

identify a plurality of data subgroups of battery charging data based on the first real-time battery charging data, wherein each data subgroup is associated with one of the battery charging modes;

receive power from a power source electrically coupled to the battery of the mobile electronic device and being configured to charge the battery of the mobile electronic device; and

in response to the battery of the mobile electronic device achieving a maximum charge:

train, using the general battery-health model as a starting model, a first customized battery-health model in a cascaded workflow using each of the plurality of data subgroups of battery charging data for the plurality of battery charging modes, wherein the first customized battery-health model is a second deep neural network trained to calculate state-of-health values associated with the battery of the mobile electronic device for the plurality of battery charging modes, respectively;

generate, using the first customized battery-health model, a plurality of first state-of-health values associated with the battery of the mobile electronic device for the plurality of battery charging modes, respectively, based on real-time battery output capacity, real-time battery output voltage, and real-time battery output current, wherein the generated plurality of first state-of-health values enable a battery management system of the mobile electronic device to prolong a battery life of the battery of the mobile electronic device; and

alleviate a potential unexpected deactivation of the mobile electronic device based at least in part on the generated plurality of first state-of-health values.

10 . The mobile electronic device of claim 9 , wherein the processors are further configured to execute the instructions to:

determine a safety risk associated with the mobile electronic device based on the plurality of first state-of-health values associated with the plurality of battery charging modes; and

present, by the mobile electronic device, an indication of the determined safety risk and the plurality of first state-of-health values associated with the mobile electronic device for the plurality of battery charging modes.

11 . The mobile electronic device of claim 10 , wherein the processors are further configured to execute the instructions to:

generate a work order of maintenance based on the safety risk associated with the mobile electronic device and the plurality of first state-of-health values associated with the mobile electronic device for the plurality of battery charging modes; and

send, to the remote server, the work order of maintenance for the mobile electronic device.

12 . The mobile electronic device of claim 9 , wherein the processors are further configured to execute the instructions to:

access the historic battery output capacity, historic battery output voltage, and historic output current associated with the mobile electronic device;

train the general battery-health model based on the historic battery output capacity, historic battery output voltage and historic battery output current associated with the mobile electronic device; and

send, from the remote server to the mobile electronic device, the general battery-health model via a wireless network.

13 . The mobile electronic device of claim 9 , wherein the processors are further configured to execute the instructions to:

access second real-time battery charging data from the mobile electronic device from a latest charging cycle; wherein the second real-time battery charging data comprises battery input voltage and battery input current during each of the plurality of battery charging modes from the latest charging cycle;

train, using the first customized battery-health model as a starting model, a second customized battery-health model using the second real-time battery charging data from the mobile electronic device from the latest charging cycle; and

determine, using the second customized battery-health model, a plurality of second state- of-health values associated with the mobile electronic device for the plurality of battery charging modes, respectively, based on the real-time battery output capacity, the real-time battery output voltage, and the real-time battery output current.

14 . The mobile electronic device of claim 9 ,

wherein the plurality of battery charging modes includes a fast charging mode and a slow charging mode, and

wherein the general battery-health model and customized battery-health model are based on a lithium-ion battery associated with the mobile electronic device.

15 . The mobile electronic device of claim 14 ,

wherein the real-time battery charging data has a first sampling rate of the fast charging mode and a second sampling rate of the slow charging mode, wherein the first sampling rate is at a higher frequency than the second sampling rate, and

wherein the processors are further configured to execute the instructions to:

apply over-sampling to the real-time battery charging data for the slow charging mode to match a sampling rate of the battery charging data for the fast charging mode.

16 . The mobile electronic device of claim 14 ,

wherein the real-time battery charging data has a first sampling rate of the slow charging mode and a second sampling rate of the fast charging mode, wherein the first sampling rate is at a higher frequency than the second sampling rate, and

wherein the processors are further configured to execute the instructions to:

apply over-sampling to the real-time battery charging data for the fast charging mode to match a sampling rate of the battery charging data for the slow charging mode.

17 . A computer-readable non-transitory storage media comprising instructions executable by a processor to:

access, from a remote server, a general battery-health model corresponding to a mobile electronic device, wherein the general battery-health model is a first deep neural network trained to calculate a state-of-health value for a battery of the mobile electronic device based on historic battery output capacity, historic battery output voltage, and historic battery output current associated with the mobile electronic device;

access real-time battery charging data from the mobile electronic device for a plurality of battery charging modes, wherein the real-time battery charging data comprises battery input voltage and battery input current during each of the plurality of battery charging modes;

identify a plurality of data subgroups of battery charging data based on the real-time battery charging data, wherein each data subgroup is associated with one of the battery charging modes;

train, using the general battery-health model as a starting model, a first customized battery-health model in a cascaded workflow using each of the plurality of data subgroups of battery charging data for the plurality of battery charging modes, wherein the first customized battery-health model is a second deep neural network trained to calculate state-of-health values associated with the mobile electronic device for the plurality of battery charging modes, respectively; and

determine, using the first customized battery-health model, a plurality of first state-of-health values associated with the mobile electronic device for the plurality of battery charging modes, respectively, based on real-time battery output capacity, real-time battery output voltage, and real-time battery output current;

access, from a remote server, a general battery-heath model corresponding to the mobile electronic device, wherein the general battery-health model is a first deep neural network trained to calculate a state-of-health value for a battery of the mobile electronic device based on historic battery output capacity, historic battery output voltage, and historic battery output current associated with the mobile electronic device;

access first real-time battery charging data from the mobile electronic device for a plurality of battery charging modes, wherein the first real-time battery charging data comprises battery input voltage and battery input current during each of the plurality of battery charging modes;

identify a plurality of data subgroups of battery charging data based on the first real-time battery charging data, wherein each data subgroup is associated with one of the battery charging modes;

receive power from a power source electrically coupled to the battery of the mobile electronic device and being configured to charge the battery of the mobile electronic device; and

in response to the battery of the mobile electronic device achieving a maximum charge:

train, using the general battery-health model as a starting model, a first customized battery-health model in a cascaded workflow using each of the plurality of data subgroups of battery charging data for the plurality of battery charging modes, wherein the first customized battery-health model is a second deep neural network trained to calculate state-of-health values associated with the battery of the mobile electronic device for the plurality of battery charging modes, respectively;

generate, using the first customized battery-health model, a plurality of first state-of-health values associated with the battery of the mobile electronic device for the plurality of battery charging modes, respectively, based on real-time battery output capacity, real-time battery output voltage, and real-time battery output current, wherein the generated plurality of first state-of-health values enable a battery management system of the mobile electronic device to prolong a battery life of the battery of the mobile electronic device; and

alleviate a potential unexpected deactivation of the mobile electronic device based at least in part on the generated plurality of first state-of-health values.

18 . The media of claim 17 , wherein the instructions are further executable by the processor to:

determine a safety risk associated with the mobile electronic device based on the plurality of first state-of-health values associated with the plurality of battery charging modes; and

present, by the mobile electronic device, an indication of the determined safety risk and the plurality of first state-of-health values associated with the mobile electronic device for the plurality of battery charging modes.

19 . The media of claim 18 , wherein the instructions are further executable by the processor to:

generate a work order of maintenance based on the safety risk associated with the mobile electronic device and the plurality of first state-of-health values associated with the mobile electronic device for the plurality of battery charging modes; and

send, to the remote server, the work order of maintenance for the mobile electronic device.

20 . The media of claim 17 , wherein the instructions are further executable by the processor to:

access the historic battery output capacity, historic battery output voltage, and historic output current associated with the mobile electronic device;

train the general battery-health model based on the historic battery output capacity, historic battery output voltage, and historic battery output current associated with the mobile electronic device; and

send, from the remote server to the mobile electronic device, the general battery-heath model via a wireless network.