Battery capacity estimation method and battery capacity estimation device
A battery capacity estimation method includes a first step and a second step. The first step is a step of obtaining image data of a Nyquist diagram drawn by a predetermined method, based on a Nyquist plot obtained by a predetermined AC-IR measurement. The second step is a step of obtaining a battery capacity estimate value of a battery to be measured by inputting the image data of the Nyquist diagram obtained in the first step into an input layer of a pre-trained neural network model.
1. A battery capacity estimation method comprising:
a first step of obtaining image data of a Nyquist diagram drawn by a predetermined method, based on a Nyquist plot obtained by a predetermined AC-IR measurement; and
a second step of obtaining a battery capacity estimate value of a battery to be measured, by inputting the image data of the Nyquist diagram obtained in the first step into a pre-trained neural network model, wherein:
the neural network model comprises:
an input layer configured to receive the image data of the Nyquist diagram of the battery to be measured;
an intermediate layer configured to obtain a battery capacity estimate value according to the image data input to the input layer; and
an output layer configured to output the battery capacity estimate value obtained by the intermediate layer.
2. The method according to claim 1 , wherein:
the intermediate layer includes a plurality of artificial neurons, each of the artificial neurons including at least one weighting factor indicating a strength of a connection between artificial neurons, and the intermediate layer is configured to perform a computation based on the weighting factor to obtain the battery capacity estimate value from the image data of the Nyquist diagram that are input to the input layer of the neural network model; and
the weighting factor is determined by, for the battery to be measured, learning training data including the image data of the Nyquist diagram drawn by the predetermined method based on the Nyquist plot obtained by the AC-IR measurement and an assessment value of battery capacity that is obtained when the image data of the Nyquist diagram is obtained, the image data and the assessment value being associated with each other.
3. The method according to claim 2 , wherein the Nyquist diagram in the image data contained in the training data is drawn to be sufficiently wide to contain the tolerance of the Nyquist plot obtained by the AC-IR measurement.
4. The method according to claim 3 , wherein the Nyquist diagram in the image data contained in the training data is drawn to have a width less than or equal to two times the tolerance of the Nyquist plot obtained by the AC-IR measurement.
5. The method according to claim 1 , wherein the Nyquist diagram in the image data obtained in the first step is drawn to be sufficiently wide to contain the tolerance of the Nyquist plot obtained by the AC-IR measurement.
6. The method according to claim 5 , wherein the Nyquist diagram in the image data obtained in the first step is drawn to have a width less than or equal to two times the tolerance of the Nyquist plot obtained by the AC-IR measurement.
7. A battery capacity estimation device comprising:
an input layer configured to receive image data of a Nyquist diagram of a battery to be measured;
an intermediate layer configured to obtain a battery capacity estimate value by performing a computation based on a weighting factor of a pre-trained neural network model based on the image data input to the input layer; and
an output layer configured to output the battery capacity estimate value obtained by the intermediate layer.
8. The battery capacity estimation device according to claim 7 , wherein:
the intermediate layer includes a plurality of artificial neurons, each of the artificial neurons including at least one weighting factor indicating a strength of a connection between artificial neurons, and the intermediate layer is configured to perform a computation based on the weighting factor to obtain the battery capacity estimate value from the image data of the Nyquist diagram that are input to the input layer of the neural network model; and
the weighting factor is determined by, for the battery to be measured, learning training data including the image data of the Nyquist diagram drawn by the predetermined method based on the Nyquist plot obtained by the AC-IR measurement and an assessment value of battery capacity that is obtained when the image data of the Nyquist diagram is obtained, the image data and the assessment value being associated with each other.
9. The battery capacity estimation device according to claim 8 , wherein the Nyquist diagram in the image data contained in the training data is drawn to be sufficiently wide to contain the tolerance of the Nyquist plot obtained by the AC-IR measurement.
10. The battery capacity estimation device according to claim 9 , wherein the Nyquist diagram in the image data contained in the training data is drawn to have a width less than or equal to two times the tolerance of the Nyquist plot obtained by the AC-IR measurement.