Power prediction system, power prediction device, power prediction method, program, and storage medium

1. A power prediction system comprising:

a battery removably mounted on an electric power device using electric power;

a charging device configured to charge the battery; and

a power prediction device configured to predict an amount of electric power capable of being supplied by the charging device to outside of the charging device through machine learning based on usage information indicating at least one of a usage state and a usage environment of the charging device,

wherein the power prediction device comprises a processor configured to execute instructions to:

acquire the usage information of the charging device at a first point in time;

input the acquired usage information at the first point in time to a model, the model having been learned to output the amount of electric power when the usage information of the charging device is input; and

predict the amount of electric power capable of being supplied by the charging device to the outside of the charging device at the first point in time based on an output result of the model with the acquired usage information at the first point in time input.

2. The power prediction system according to claim 1 , wherein the processor is further configured to execute instructions to:

derive the amount of electric power capable of being supplied by the charging device to the outside of the charging device based on the number of batteries mounted on the charging device and a charging capacity of the battery mounted on the charging device, and

learn the model based on a dataset in which the derived amount of electric power is associated with the usage information of the charging device.

3. The power prediction system according to claim 2 ,

wherein the processor is further configured to execute instructions to:

acquire the usage information of the charging device for a region where each of a plurality of charging devices is installed, and

learn the model for each region based on the usage information acquired from the charging device for each region.

4. The power prediction system according to claim 1 , wherein the processor is further configured to execute instructions to: predict the amount of electric power based on the model when the charging device has received a limited power consumption request.

5. The power prediction system according to claim 4 , wherein the processor is further configured to execute instructions to: cause electric power to be supplied from the charging device to the outside of the charging device if the predicted amount of electric power is greater than or equal to an amount of electric power of a limited power consumption request when the charging device has received the limited power consumption request.

6. The power prediction system according to claim 1 , wherein the usage state comprises at least one of an index value indicating a charging capacity of the battery mounted on the charging device, the number of batteries mounted on the charging device, and a time.

7. The power prediction system according to claim 1 , wherein the usage environment comprises at least one of weather of a region where the charging device is installed, a temperature of the region where the charging device is installed, a date, and a day of the week.

8. A power prediction device comprising:

a receiver configured to receive usage information indicating at least one of a usage state and a usage environment of a charging device transmitted from the charging device configured to charge a battery removably mounted on an electric power device using electric power; and

a processor configured to execute instructions to: predict an amount of electric power capable of being supplied by the charging device to outside of the charging device through machine learning based on the usage information of the charging device received by the receiver,

wherein the receiver configured to receive the usage information of the charging device at a first point in time; and

wherein the processor is further configured to further execute instructions to:

input the received usage information at the first point in time to a model, the model having been learned to output the amount of electric power when the usage information of the charging device is input, and

predict the amount of electric power capable of being supplied by the charging device to the outside of the charging device at the first point in time based on an output result of the model with the received usage information at the first point in time input.

9. A power prediction method comprising steps of:

receiving, by a computer, usage information indicating at least one of a usage state and a usage environment of a charging device transmitted from the charging device configured to charge a battery removably mounted on an electric power device using electric power;

predicting, by the computer, an amount of electric power capable of being supplied by the charging device to outside of the charging device through machine learning based on the received usage information;

receiving, by the computer, the usage information of the charging device at a first point in time;

inputting, by the computer, the received usage information at the first point in time to a model, the model having been learned to output the amount of electric power when the usage information of the charging device is input; and

predicting, by the computer, the amount of electric power capable of being supplied by the charging device to the outside of the charging device at the first point in time based on an output result of the model with the received usage information at the first point in time input.

10. A computer-readable non-transitory storage medium storing a program for causing a computer to execute steps of:

receiving usage information indicating at least one of a usage state and a usage environment of a charging device transmitted from the charging device configured to charge a battery removably mounted on an electric power device using electric power; and

predicting an amount of electric power capable of being supplied by the charging device to outside of the charging device through machine learning based on the received usage information;

receiving the usage information of the charging device at a first point in time;

inputting the received usage information at the first point in time to a model, the model having been learned to output the amount of electric power when the usage information of the charging device is input; and

predicting the amount of electric power capable of being supplied by the charging device to the outside of the charging device at the first point in time based on an output result of the model with the received usage information at the first point in time input.