Vehicle battery abnormity identification method and device

The invention provides a vehicle battery abnormity identification method and device. The method comprises the following steps: determining or calculating a plurality of characteristics of each piece of battery data; calculating a covariance matrix according to the plurality of characteristics of each piece of battery data, and deleting battery data with low correlation with other battery data according to the covariance matrix; a label of each piece of battery data is determined according to the abnormity occurrence moment, one piece of battery data carrying the label is used as a training sample, a training sample set is obtained, and the label of one piece of battery data is whether the piece of battery data is abnormal or not and the abnormity category; training a random forest model by using training samples in the training sample set through a method of calculating a Gini coefficient and a Gini index; and using the trained random forest model to identify whether the vehicle battery is abnormal or not and