A Novel Screening Approach Based on Neural Network for the Second Usage of Retired Lithiumion Batteries

With a big mass of lithium-ion batteries retired from electric vehicles, researches on battery secondary use are rather urgent and important. However, existing screening researches cannot ensure high accuracy and high efficiency meanwhile. To solve the shortcomings, an accurate and fast screening approach on the basis of Neural Network is proposed in this work. The specific partial charging curves of battery cells are decided by incremental capacity analysis for its ability to detect battery aging state. Then critical features extraction is executed based on partial charging curves. They are used as the input of the screening Neural Network model and help finish the model training. The results on 234 LiFePO4 battery cells proves the validity of the method. The screening accuracy can be high as 97.4%. As batteries count climbed to 5000, the screening efficiency can be 9 times higher than traditional capacity testing approach.