Researchers Develop Novel Deep Learning Model to Predict Battery Lifetime
The deep learning model effectively eliminated the dependence on a large amount of charging test data and provided a new idea for predicting battery life in real-time, the journal IEEE Transactions on Transportation Electrification reported.
The article noted that the accurate lifetime prediction of LIBs is essential to the normal and effective operation of electric devices. However, such estimation faces huge challenges due to the nonlinear capacity degradation process and uncertain operating conditions of LIBs.
The researchers from the Dalian Institute of Chemical Physics (DICP) of the Chinese Academy of Sciences and the Xi'an Jiaotong University proposed a deep learning model based on a small amount of charge cycle data to predict the target battery's current cycle life and remaining useful life.
The learning model can accurately predict the battery's current cycle life and remaining service life using only 15 charge cycle data. According to the experiment results, this data can make an accurate prediction.
The proposed model is expected to provide a solution for intelligent battery management, said Chen Zhongwei, the director of the State Key Laboratory of Catalysis, DICP.
4155/v
