# Automated feature extraction and selection for data-driven models of rapid battery capacity fade and end of life

arXiv: https://arxiv.org/abs/2101.04440 · HTML: https://mlsanity.com/paper/2101.04440/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Jan 2021.

## Built on by (most cited first, 9 of 9)

- [Predicting Battery Lifetime Under Varying Usage Conditions from Early Aging Data](https://mlsanity.com/paper/2307.08382/) (2023)
- [Transfer Learning-based State of Health Estimation for Lithium-ion Battery with Cycle Synchronization](https://mlsanity.com/paper/2208.11204/) (2022)
- [Cycle Life Prediction for Lithium-ion Batteries: Machine Learning and More](https://mlsanity.com/paper/2404.04049/) (2024)
- [Evaluating feasibility of batteries for second-life applications using machine learning](https://mlsanity.com/paper/2203.04249/) (2022)
- [Systematic Feature Design for Cycle Life Prediction of Lithium-Ion Batteries During Formation](https://mlsanity.com/paper/2410.07458/) (2024)
- [Learning battery model parameter dynamics from data with recursive Gaussian process regression](https://mlsanity.com/paper/2304.13666/) (2023)
- [Foundation Models Knowledge Distillation For Battery Capacity Degradation Forecast](https://mlsanity.com/paper/2505.08151/) (2025)
- [Lithium-ion Battery Online Knee Onset Detection by Matrix Profile](https://mlsanity.com/paper/2304.00691/) (2023)
- [Prognosis of Multivariate Battery State of Performance and Health via Transformers](https://mlsanity.com/paper/2309.10014/) (2023)

## Abstract (arXiv)

Lithium-ion cells may experience rapid degradation in later life, especially with more extreme usage protocols. The onset of rapid degradation is called the `knee point', and forecasting it is important for the safe and economically viable use for batteries. We propose a data-driven method that uses automated feature selection to produce inputs for a Gaussian process regression model that estimates changes in battery health, from which the entire capacity fade trajectory, knee point and end of life may be predicted. The feature selection procedure flexibly adapts to varying inputs and prioritises those that impact degradation. For the datasets considered, it was found that calendar time and time spent in specific voltage regions had a strong impact on degradation rate. The approach produced median root mean square errors on capacity estimates under 1%, and also produced median knee point and end of life prediction errors of 2.6% and 1.3% respectively.