Battery life learning device, battery life prediction device, method and non-transitory computer readable medium
A battery life learning device including a learning section configured to obtain a learned prediction model for predicting a remaining life of a vehicle battery from time-series data of a deterioration characteristic of the vehicle battery, the learned prediction model being obtained by learning a prediction model from the time-series data of the deterioration characteristic of the vehicle battery based on learning data, the learning data including time-series data of the deterioration characteristic at a predetermined time point in the past of a vehicle battery for learning that has reached an end of life and the remaining life at the predetermined time point of the vehicle battery for learning.
1. A battery life learning device, comprising:
a learning section configured to obtain a learned prediction model for predicting a remaining life of a vehicle battery from time-series data of a deterioration characteristic of the vehicle battery, the learned prediction model being obtained by learning a prediction model from the time-series data of the deterioration characteristic of the vehicle battery based on learning data, the learning data including time-series data of the deterioration characteristic at a predetermined time point in the past of a vehicle battery for learning that has reached an end of life and a remaining life at the predetermined time point of the vehicle battery for learning.
2. The battery life learning device of claim 1 , further comprising a generation section configured to generate a plurality of partial time-series data as the learning data, the plurality of partial time-series data being generated by:
acquiring the time-series data of the deterioration characteristic of the vehicle battery for learning,
dividing the acquired time-series data for each predetermined period, and
assigning the remaining life of the vehicle battery for learning as a label for each divided partial time-series data.
3. The battery life learning device of claim 1 , wherein the learned prediction model includes a probabilistic remaining life determination based on a weight value and a bias value.
4. A battery life prediction device, comprising:
an acquisition section configured to acquire time-series data of a deterioration characteristic of a prediction target vehicle battery; and
a prediction section configured to predict a remaining life of the prediction target vehicle battery based on the time-series data of the deterioration characteristic of the prediction target vehicle battery acquired by the acquisition section and based on a learned prediction model that has been learned in advance from learning data, the learning data including time-series data of the deterioration characteristic at a predetermined time point in the past of a vehicle battery for learning that has reached an end of life and a remaining life at the predetermined time point of the vehicle battery for learning.
5. The battery life prediction device of claim 4 , wherein the prediction section is configured to stop the prediction of the remaining life of the prediction target vehicle battery in a case in which an appearance frequency of an unmeasured portion in the deterioration characteristic is equal to or greater than a predetermined value or more in the time-series data of the deterioration characteristic of the prediction target vehicle battery.
6. The battery life prediction device of claim 4 , further comprising a display control section configured to cause a display section to display the time-series data of the deterioration characteristic of the prediction target vehicle battery, and to display comments corresponding to the remaining life of the prediction target vehicle battery predicted by the prediction section.
7. The battery life prediction device of claim 4 , wherein the learned prediction model includes a probabilistic remaining life determination based on a weight value and a bias value.
8. A battery life learning method executed by a computer, the method comprising:
obtaining learned prediction model for predicting a remaining life of a vehicle battery from time-series data of a deterioration characteristic of the vehicle battery, the learned prediction model being obtained by learning a prediction model from the time-series data of the deterioration characteristic of the vehicle battery based on learning data, the learning data including time-series data of the deterioration characteristic at a predetermined time point in the past of a vehicle battery for learning that has reached an end of life and a remaining life at the predetermined time point of the vehicle battery for learning.
9. The battery life learning method of claim 8 , wherein the learned prediction model includes a probabilistic remaining life determination based on a weight value and a bias value.
10. A battery life prediction method executed by a computer, the method comprising:
acquiring time-series data of a deterioration characteristic of a prediction target vehicle battery; and
predicting a remaining life of the prediction target vehicle battery based on the time-series data of the deterioration characteristic of the prediction target vehicle battery acquired by the acquisition section and based on a learned prediction model that has been learned in advance from learning data, the learning data including time-series data of the deterioration characteristic at a predetermined time point in the past of a vehicle battery for learning that has reached an end of life and a remaining life at the predetermined time point of the vehicle battery for learning.
11. The battery life prediction method of claim 10 , wherein the learned prediction model includes a probabilistic remaining life determination based on a weight value and a bias value.
12. A non-transitory computer readable medium storing a program causing a computer to execute processing for battery life learning, the processing comprising:
obtaining a learned prediction model for predicting a remaining life of a vehicle battery from time-series data of a deterioration characteristic of the vehicle battery, the learned prediction model obtained by learning a prediction model from the time-series data of the deterioration characteristic of the vehicle battery based on learning data, the learning data including time-series data of the deterioration characteristic at a predetermined time point in the past of a vehicle battery for learning that has reached an end of life and a remaining life at the predetermined time point of the vehicle battery for learning.
13. The non-transitory computer readable medium of claim 12 , wherein the learned prediction model includes a probabilistic remaining life determination based on a weight value and a bias value.
14. A non-transitory computer readable medium storing a program causing a computer to execute processing for battery life prediction, the processing comprising:
acquiring time-series data of a deterioration characteristic of a prediction target vehicle battery; and
predicting a remaining life of the prediction target vehicle battery based on the time-series data of the deterioration characteristic of the prediction target vehicle battery acquired by the acquisition section and based on a learned prediction model that has been learned in advance from learning data, the learning data including time-series data of the deterioration characteristic at a predetermined time point in the past of a vehicle battery for learning that has reached an end of life and a remaining life at the predetermined time point of the vehicle battery for learning.
15. The non-transitory computer readable medium of claim 14 , wherein the learned prediction model includes a probabilistic remaining life determination based on a weight value and a bias value.