An Improved Data-Driven Life Prediction Model of Lithium-Ion Battery

Due to high energy density, low self-discharge rate and long cycle life, lithium-ion battery has gained more and more popularities in this modern age. However, its safety has gradually become a major concern, so accurate life prediction is essential to ensure its reliability. Considering the influence of calendar aging on cycle life, this paper proposes an easy-to-obtain calendar aging feature, which can be combined with cycle aging features to formulate the life prediction model. With the open-source experimental data, the proposed calendar aging feature is verified to be effective in accurate life prediction. Ensured by good accuracy, the formulated life prediction model can be integrated into battery management system, and serve as a practical tool to enhance the reliability of battery packs.