Abstract：Lithium-ion batteries are used as energy storage system components for electric vehicles, unmanned aerial vehicles and power electronic equipment. Accurate state of charge (SOC) estimation plays an important role in correct decision-making, safe control and maintenance. Aiming at the problem of lithium-ion battery SOC estimation, this paper uses long short-term memory (LSTM) neural network to build a lithium-ion battery SOC prediction model, taking battery voltage, current and temperature as inputs, building a multi-layer LSTM prediction model and adopting Adam optimization algorithm. The Adam optimization algorithm complete the training of the LSTM model. The verification of training and test results shows that adding Adam algorithm and Dropout regularization method to the model training process is robust to the nonlinearity of the experimental data set and the uncertainty of the initial SOC.
潘锦业, 王苗苗, 阚威, 高永峰. 基于Adam优化算法和长短期记忆神经网络的锂离子电池荷电状态估计方法[J]. 电气技术, 2022, 23(4): 25-30.
PAN Jinye, WANG Miaomiao, KAN Wei, GAO Yongfeng. State of charge estimation of lithium-ion battery based on Adam optimization algorithm and long short-term memory neural network. Electrical Engineering, 2022, 23(4): 25-30.
2022, Vol. 23 
