Piecewise-linear modelling with feature selection for Li-ion battery end of life prognosis

TL;DR

This paper presents a fast, flexible piecewise-linear modeling approach with automated feature selection for lithium-ion battery end-of-life prognosis, offering comparable accuracy to Gaussian process regression and robustness to data variability.

Contribution

The study introduces a novel piecewise-linear modeling method combined with automated feature selection for battery health forecasting, balancing speed and flexibility.

Findings

Piecewise-linear models perform as well as Gaussian process regression.

Automated feature selection improves model performance.

Method is robust to varying input size and data availability.

Abstract

The complex nature of lithium-ion battery degradation has led to many machine learning based approaches to health forecasting being proposed in literature. However, machine learning can be computationally intensive. Linear approaches are faster but have previously been too inflexible for successful prognosis. For both techniques, the choice and quality of the inputs is a limiting factor of performance. Piecewise-linear models, combined with automated feature selection, offer a fast and flexible alternative without being as computationally intensive as machine learning. Here, a piecewise-linear approach to battery health forecasting was compared to a Gaussian process regression tool and found to perform equally well. The input feature selection process demonstrated the benefit of limiting the correlation between inputs. Further trials found that the piecewise-linear approach was robust to changing input size and availability of training data.