Electrochemical Impedance Spectroscopy Based on the State of Health Estimation for Lithium-Ion Batteries

The state of health (SOH) is critical to the efficient and reliable use lithium-ion batteries (LIBs). Recently, SOH estimation method based on electrochemical impedance spectroscopy (EIS) has been proven effective. In response different practical applications, two models for are proposed in this paper. Aiming at equivalent circuit model (ECM) method, a variety ECMs proposed. Used EIS predict ECM, results show that improved ensures correctness ECM improves SOH. data-driven algorithm, proposes convolution neural network (CNN) process data which can not only extract key points but also simplifies complexity manual feature extraction. bidirectional long short-term memory (BiLSTM) was used serial regression prediction. Moreover, Particle Swarm Optimization (IPSO) algorithm optimize model. Comparing (IPSO-CNN-BiLSTM) with traditional PSO-CNN-BiLSTM, CNN-BiLSTM LSTM models, prediction by 13.6%, 93.75% 94.8%, respectively. Besides that, methods 27% 35% better than existing gaussion (GPR) model, indicates more flexible higher precision.