Remaining useful life prediction of LiFePO 4 battery based on particle filter

LiFePO 4 battery has been widely used in electric vehicles due to high safety and long cycle life. This paper firstly analyses the basic characteristics of LiFePO 4 battery including the capacity and internal resistance. Secondly, particle filter (PF) algorithm is introduced to predict the remaining useful life (RUL) of LiFePO 4 battery effectively. Based on the LiFePO 4 battery life degradation data, the prediction accuracy of four kinds of resampling algorithm is analyzed and compared. The result of Systematic Resampling is the most close to real life end points, and Random Resampling has the lowest prediction accuracy. Therefore, Systematic Resampling is used to predict RUL. The results indicate that PF can efficiently predict RUL of LiFePO 4 Battery.