Battery inconsistency judgment method and device

Abstract

The invention relates to the technical field of new energy vehicles, and discloses a battery inconsistency judgment method and device, and the method comprises the following steps: S1, obtaining SOC data of a vehicle at the tail end of a charging working condition in different charging states; s2, on the basis of the SOC data, calculating and linearly fitting the voltage difference of each frame under the charging condition for multiple times, so as to obtain a trend feature describing the trend of the voltage difference along with the time change; and S3, carrying out statistics on a linear fitting result and the voltage difference in the multiple charging working conditions to form a discrete degree feature for describing the discrete degree of the voltage difference relative to the trend. Parameter fitting is carried out on the voltage difference value of the charging tail end, the fitting slope, the fitting goodness, the mean value of the voltage difference, the maximum value of the voltage difference and other characteristic quantities are extracted, machine learning training is used for obtaining the weight parameters of all the characteristics, synchronous input of multiple characteristics is achieved, misjudgment caused by a single characteristic is avoided, and the judgment accuracy is improved.

Images