Battery capacity estimation method and battery capacity estimation system

Abstract

A battery capacity estimation system executes a charging and discharging process (S1), an alternating current impedance acquiring process (S2 and S3), and a battery capacity estimating process (S4 to S6). The charging and discharging process involves charging and discharging a target secondary battery. The alternating current impedance acquiring process involves acquiring a measurement result of an alternating current impedance of a target secondary battery, by applying an alternating current signal within a specific frequency range to the target secondary battery after completion of the charging and discharging in the charging and discharging process and before a predetermined maximum waiting time elapses. The battery capacity estimating step involves estimating a battery capacity of the target secondary battery based on the measurement result of the alternating current impedance.

What is claimed is: 
     
         1 . A battery capacity estimation method comprising:
 a charging and discharging step of charging and discharging a target secondary battery;   an alternating current impedance acquiring step of acquiring a measurement result of an alternating current impedance of the target secondary battery, by applying an alternating current signal within a specific frequency range to the target secondary battery after completion of the charging and discharging in the charging and discharging step and before a predetermined maximum waiting time elapses; and   a battery capacity estimating step of estimating a battery capacity of the target secondary battery based on the measurement result of the alternating current impedance, wherein:   the battery capacity estimating step includes obtaining an estimation result of the battery capacity of the target secondary battery by inputting data based on a Nyquist plot of the target secondary battery into a pre-trained neural network model; and   the pre-trained neural network model performs learning using a plurality of training data including data based on Nyquist plots indicating measurement results of alternating current impedance of a plurality of secondary batteries and actual battery capacities of the plurality of secondary batteries.   
     
     
         2 . The battery capacity estimation method according to  claim 1 , wherein the pre-trained neural network model is trained based on the training data including data relating to the Nyquist plots of the secondary batteries after completion of the charging and discharging and before the maximum waiting time elapses.