Prediction of lithium-ion battery capacity by functional monitoring data using functional principal component analysis

Abstract

Lithium-ion batteries have been a promising energy storage technology for applications such as electronics, automobiles, and smart grids over the years. Extensive research was conducted to improve the prediction of the remaining capacity of the lithium-ion battery. A robust prediction model would improve the battery performance and reliability for forthcoming usage. To develop a data-driven capacity prediction model of lithium-ion batteries most of past studies employed capacity degradation data, yet very few tried using other performance monitoring variables such as temperature, voltage, and current data to estimate and predict the battery capacity. In this thesis, we aim to develop a data-driven model for predicting the capacity of lithium-ion battery adopting functional principal component analysis applied to functional monitoring data of temperature, voltage, and current observations collected from NASA Ames Prognostics Center of Excellence repository. The result of capacity prediction has been substantiated with past studies and obtained root mean square error (RMSE) of 0.009. The proposed data-driven approach performs well to predict the capacity employing functional performance measures over the life span of a lithium-ion battery.