Prediction of State of Charge of Lithium Ion Battery Based on CNN-LSTM
In the battery management system,state of charge(SOC)is the key parameter of lithium-ion battery,and its estimation accuracy is particularly important for the battery management system.A data drive method based on CNN-LSTM network is proposed.The voltage,current and temperature of the battery are selected as the input and SOC is selected as the output.The CNN-LSTM network is used to extract the nonlinear correlation,spatial property and temporal property between input and output to accurately predict SOC.Meanwhile,the network parameters are determined by combination method.Finally,the experiment is carried out by MATLAB software.The experimental re-sults show that RMSE is still lower than 2%at low temperature,which has high accuracy and broad application prospect.
lithium-ion batterystate of chargeconvolutional neural networklong and short term neural networks
NETL
NSTL
万方数据
