Lithium-Ion Battery Health Prognosis Based on a Real Battery Management System Used in Electric Vehicles

TLDR

The study develops a health indicator based on partial charge voltage curves and a moving‑window method to predict lithium‑ion battery remaining useful life. The indicator is extracted from partial charge voltage data, and remaining useful life is forecasted with a linear aging model built on capacity measurements within a moving window, augmented by Monte Carlo simulation, all implemented on a real vehicle battery management system and validated with cells tested at 1–2 C and 25–40 °C. Capacity estimation errors were below 1.5 %, and during the final 20 % of life the root‑mean‑square prediction error was within 20 cycles, with 95 % confidence intervals covering roughly the same range.

Abstract

This paper developed an effective health indicator to indicate lithium-ion battery state of health and moving-window-based method to predict battery remaining useful life. The health indicator was extracted based on the partial charge voltage curve of cells. Battery remaining useful life was predicted using a linear aging model constructed based on the capacity data within a moving window, combined with Monte Carlo simulation to generate prediction uncertainties. Both the developed capacity estimation and remaining useful life prediction methods were implemented based on a real battery management system used in electric vehicles. Experimental data for cells tested at different current rates, including 1 and 2 C, and different temperatures, including 25 and 40 °C, were collected and used. The implementation results show that the capacity estimation errors were within 1.5%. During the last 20% of battery lifetime, the root-mean-square errors of remaining useful life predictions were within 20 cycles, and the 95% confidence intervals mainly cover about 20 cycles.