A Data-Driven Remaining Useful Life Prediction Approach for Lithium-Ion Batteries Based on Charging Health Feature Optimization
The Remaining Useful Life(RUL)prediction accuracy of lithium battery is not high because the selected health factors are not ideal.To solve this problem,this paper proposed a data-driven remaining useful life estimation approach for lithium-ion batteries based on charging health feature optimization.Firstly different health factors were selected in the battery charging process,then,a two-step feature selection method based on maximum information coefficient was used to obtain optimal health factors.Finally,the Attention Temporal Convolutional Network(ATCN)mechanism was used to predict the remaining useful life of the battery.The proposed lithium battery RUL prediction framework was validated by a study of NASA's lithium battery aging data and compared with other modeling methods including Simple Recurrent Neutral Network(SimpleRNN),Long Short Term Memory(LSTM)neutral network and Gate Recurrent Unit(GRU)neutral network.The experimental results indicate the proposed method has achieved optimal prediction results in all the datasets.
万方数据
维普
