Estimation of state-of-charge and state-of-health for lithium-ion battery based on improved firefly optimized particle filter

Abstract

Accurate estimation of state-of-charge and state-of-health is extremely important for the lithium-ion batteries used in electric vehicles. First, this article proposes an online parameter identification method based on joint forgetting factor recursive least squares-adaptive extended Kalman filter and calculate state-of-charge by adaptive extended Kalman filter in real time. Through simulation tests, the minimum error of this method is 0.45%, which verifies its high accuracy. Then an improved firefly algorithm solved the particle dilution problem is combined with particle filter to estimate state-of-health for the first time. Compared with particle filter, the estimation accuracy of improved firefly algorithm optimized particle filter is better in the experiment, its estimated error is as low as 0.79% and predicted cycles is consistent with the failure cycles in test. In addition, the convergence of the algorithm is verified by the experiment.