Remaining Useful Life Prediction of Lithium-Ion Battery Based on Gauss–Hermite Particle Filter

This brief proposes a prediction method of remaining useful life (RUL) based on Gauss-Hermite particle filter (GHPF) in nonlinear and non-Gaussian systems of Lithium-ion batteries (LIBs). In this brief, to improve the accuracy and reduce the computational complexity of the estimation of state of health (SOH), multiscale extended Kalman filter is proposed to execute state of charge (SOC) and SOH joint estimation with dual time scales because of the slow-varying characteristic of SOH and fast-varying characteristic of SOC. Based on the estimation of SOH, a GHPF is developed to update the parameters of the capacity degradation model in real time and predict the RUL of LIBs. The simulation results show that the proposed prediction method of RUL has a better performance and higher precision than the method based on standard PF.