Remaining Useful Life Prediction for Lithium-ion Battery Using Ensemble Learning Method

• Published in: 2019 IEEE Power & Energy Society General Meeting (PESGM) pp. 1 - 5
• Main Authors: Gou, Bin, Xu, Yan, Fang, Sidun, Pratama, Ryan Arya, Liu, Shuyong
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2019
• Subjects: ensemble learning; Lithium-ion batteries (LIBs); Random Vector Functional Link (RVFL) network; remaining useful life (RUL)
• ISSN: 1944-9933
• DOI: 10.1109/PESGM40551.2019.8973811

Lithium-ion batteries (LIBs) have been widely applied in energy storage system, and its prognostics and health management are of great importance for the system performance. In this paper, an ensemble data-driven method is proposed to accurately predict the remaining useful life (RUL) of lithium-ion batteries. A strongly related feature extracted from the charging voltage is selected to indicate the degradation trend of battery, according to the Pearson correlation analysis. A random learning algorithm named Random Vector Functional Link (RVFL) network is applied to map the knowledge relationship between the extracted feature and practical health state of the battery due to its fast learning speed. The nonlinear autoregressive with exogenous inputs (NARX) structure which contains past and present information is introduced to set up the deterioration process of the LIBs. An ensemble learning method that combines a set of single RVFL models is designed to further improve the RUL prediction accuracy. The test results using publicly available dataset show that the proposed ensemble data-driven method can accurately predict the RUL of batteries without any other additional hardware or system downtime.