An Immune Particle Swarm Self-Attention Mechanism-Long Short-Memory Neural Network Method for High-Precision State-of-Charge Evaluation of Lithium-ion Batteries

Estimating the state of charge (SOC) of lithium-ion batteries is crucial for the effective operation of battery management systems. A SOC estimation method based on the immune particle swarm-simplified self-attentive mechanism long short-term memory neural network (IPSO-SSAM-LSTM) is proposed to improve the estimation accuracy of li-ion batteries in complex environments. An immune particle swarm algorithm is used to optimize network hyperparameters and improve model accuracy automatically. The experimental results show that the IPSO-SSAM-LSTM model outperforms the other models, and the errors of the SOC prediction results are all less than 1.00%, which indicates that the IPSO-SSAM-LSTM model is able to predict the SOC accurately and with high precision.