LAUSR

Abstract In this study, a more precise numerical method to discretize the equation of State-of-Charge is proposed. Unscented particle filter and cubature Kalman filter are performed to estimate State-of-Charge. A hybrid cubature particle filter is presented by aggregating the cubature filter and particle filter to achieve a more stable estimation of State-of-Charge under harsh charging & discharging schedules. Furthermore, the noise self-adjustment strategy is applied to make the proposed estimator more applicable to practical engineering environment. Extensive experiments are conducted on the real data from the Federal Urban Driving Schedule and Dynamic Stress Test, and the results verify that the proposed hybrid method is more robust than the existing models.