State of Health (SOH) Estimation of Lithium-Ion Batteries Based on CNN-LSTM Modeling

Lithium-ion batteries are widely utilized in new energy applications such as electric vehicles and energy storage systems, owing to their high energy density, lightweight design, and long cycle life. However, capacity degradation over time—primarily caused by the loss of active materials during charge-discharge cycling—significantly affects their service life, performance reliability, and operational safety. Accurate estimation of battery state of health (SOH) is therefore essential for ensuring the safe and efficient operation of battery-powered systems. In this study, a deep learning model that integrates a convolutional neural network (CNN) with a long short-term memory (LSTM) network is proposed to predict the SOH of lithium-ion batteries. This hybrid model leverages CNN’s strength in extracting spatial features and LSTM’s capability in modeling temporal dependencies, enabling effective learning from battery cycling data. Experimental results show that the model achieves a coefficient of determination (R2) of 0.999240 on the training set and 0.950681 on the test set, indicating high estimation accuracy and strong generalization. The proposed method offers a robust and scalable solution for intelligent battery health management and lifecycle optimization.