State-of-Health prediction of lithium-ion batteries based on a low dimensional Gaussian Process Regression

• Published in: Journal of energy storage 2024-05, Vol.88, p.111649, Article 111649
• Main Authors: Pohlmann, Sebastian, Mashayekh, Ali, Stroebl, Florian, Karnehm, Dominic, Kuder, Manuel, Neve, Antje, Weyh, Thomas
• Format: Article
• Language: English
• Subjects: Gaussian Process Regression; Lithium-ion battery; Machine learning; State of health
• ISSN: 2352-152X; 2352-1538
• DOI: 10.1016/j.est.2024.111649

An accurate determination of the condition of a battery is a key challenge in operation. As the performance of lithium-ion batteries is degrading over time, an accurate prediction of the State-of-Health would improve the overall efficiency and safety. This paper presents a prediction method for the State-of-Health based on a Gaussian Process Regression with an automatic relevance determination kernel in a single model for three different types of battery cells. After reducing the dimension of the problem and a sensitivity analysis of the features, the model is trained, validated, and further tested on unseen data. A minimum test error is obtained with a mean absolute error of 1.33%. Combined with the low uncertainty of the prediction results, this shows the applicability and the great potential of forecasting the condition of a battery using data-driven methods. •Comparison of dimension reduction methods and sensitivity analysis of features.•Analysis of data augmentation for battery degradation data.•Selection of a suitable kernel for the Gaussian Process Regression.•Prediction of the battery State-of-Health for different battery types.•Discussion of data driven aspects to predict the condition of a battery.