Skip to Main Content
Most methods for state-of-charge or state-of-health prognostics are impedance based. Impedance models must be as simple as possible to be implemented on embedded applications but as accurate as possible to represent the main electrochemical phenomena. The parameters of such models can be identified using impedance spectroscopy. However, the electrochemical features change in relation to numerous parameters, such as temperature or aging, which involves updating the model of the battery online. This paper deals with the use of an extended Kalman filter (EKF) for the observation of the parameters of a Li-ion battery lumped model. First, this paper will focus on the electrical model that can be used to represent the main electrochemical phenomena in the battery. Then, mathematical considerations about the EKF are reminded to be applied to the observation of the impedance parameters of the battery. This method has been validated on an urban driving cycle of a hybrid electric vehicle.