State-of-health estimation of Li-ion batteries in the early phases of qualification tests: An interpretable machine learning approach

作者：

Highlights：

• Developing machine learning models for the SOH estimation of Li-ion batteries.

• Extracting three types of statistical features from capacity fade data.

• Interpreting the model’s behaviors and predictions based on SHAP.

• Reducing the time required for qualification tests to 100 cycles.

摘要

•Developing machine learning models for the SOH estimation of Li-ion batteries.•Extracting three types of statistical features from capacity fade data.•Interpreting the model’s behaviors and predictions based on SHAP.•Reducing the time required for qualification tests to 100 cycles.

论文关键词：State-of-health estimation,Qualification test,Li-ion battery,Interpretable machine learning,SHapley Additive exPlanation method

论文评审过程：Received 11 August 2021, Revised 1 March 2022, Accepted 2 March 2022, Available online 5 March 2022, Version of Record 9 March 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116817