Smoothing Method Enhances Machine-Learning-Based State-of-Health Prediction for Lithium-Ion Batteries

Written by

in

Researchers proposed a Random Forest-based state-of-health prediction framework that combines Rolling Mean and Savitzky–Golay smoothing techniques to improve SOH prediction accuracy throughout the lithium-ion battery lifecycle. The method reduces short-term measurement noise while preserving long-term degradation trends, and experimental validation shows particularly notable performance gains in the later stages of battery cycling.

Source:BIOENGINEER.ORG · Read original

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *