According to Tianyan inspection property clue information, recently, a patent for a “battery health prediction method and device” applied for by Xiaomi Auto Technology Co., Ltd. was announced. The summary shows that this disclosure relates to the field of simulation computation. The specific steps are: obtaining the first training data set, training a battery state prediction model based on the first training data set to determine the first parameter set, wherein the first training data set is battery state operation data; obtaining the second training data set, training the battery state prediction model to determine the second parameter set based on the second training data set, where the second training data set is data related to battery health status; configuring the first parameter set and the second parameter set into the battery state prediction model; and inference the real-time parameters of the battery into the battery state prediction model for inference operations to determine the battery health state parameters. The present disclosure trains parameters in the battery state prediction model through two training data sets, enables prediction of battery health state parameters, avoids the problem of insufficient prediction accuracy, and improves the accuracy of battery state prediction.

According to Tianyan inspection property clue information, recently, a patent for a “battery health prediction method and device” applied for by Xiaomi Auto Technology Co., Ltd. was announced. The summary shows that this disclosure relates to the field of simulation computation. The specific steps are: obtaining the first training data set, training a battery state prediction model based on the first training data set to determine the first parameter set, wherein the first training data set is battery state operation data; obtaining the second training data set, training the battery state prediction model to determine the second parameter set based on the second training data set, where the second training data set is data related to battery health status; configuring the first parameter set and the second parameter set into the battery state prediction model; and inference the real-time parameters of the battery into the battery state prediction model for inference operations to determine the battery health state parameters. The present disclosure trains parameters in the battery state prediction model through two training data sets, enables prediction of battery health state parameters, avoids the problem of insufficient prediction accuracy, and improves the accuracy of battery state prediction.