Protocol for state-of-health prediction of lithium-ion batteries based on machine learning

Abstract

Summary: Accurate estimates of State of Health (SoH) are critical for characterizing the aging of lithium-ion batteries. This protocol combines feature extraction and a representative machine learning algorithm (i.e., least-squares support vector machine) for SoH prediction of lithium-ion batteries. We detail the step-by-step estimation process, followed by validation of the constructed model with a maximum absolute error of 1.62%. Overall, the proposed approach can efficiently track the aging trajectory and ensure precise SoH prediction.For complete details on the use and execution of this protocol, please refer to Shu et al. (2021b).

Keywords

• Physics
• Energy
• Material sciences
• Computer sciences