State of Charge Estimation of LiFePO4 Battery Used in Electric Vehicles Using Support Vector Regression, PCA and DP Battery Model

This paper presents a new methodology for State-of-Charge (SoC) estimation based on Support Vector Regression (SVR), Principal component Analysis (PCA) and the Dual Polarization (DP) battery model. The proposed methodology considers several factors that are significant for the prediction of battery current in Electric Vehicles (EVs) such as: speed, acceleration, voltage, pedal position; this information is then used as input to the DP battery model. Battery parameters were estimated using the Nonlinear Least Square (NLS) algorithm.