Remaining Useful Life Indirect Prediction of Lithium-ion Batteries Based on Gaussian Mixture Regression

Abstract

Remaining useful life (RUL) prediction of lithium-ion batteries is one of the key technologies on prognostics and health management. Highly accurate RUL prediction of lithium-ion batteries is a prerequisite to ensure the safety and reliability for electric vehicles. To describe the accurate RUL prediction, the RUL indirect prediction framework based on Gaussian mixture regression (GMR) is proposed. Firstly, the discharging voltage and current indirect health indicators are extracted, and grey relation analysis (GRA) is used to analyze the relation with capacity. Then, to improve the RUL prediction performance, GMR method is proposed for reducing the impact of external disturbances. Finally, the proposed method is compared with existing methods. The results show that the proposed method is superior to traditional methods.