Online battery state of power prediction using PRBS and extended Kalman filter

This paper presents a hybrid battery parametrisation technique for the purpose of battery state-of-charge (SOC) and state-of-power (SOP) monitoring in real time. The proposed technique is centred around an opportunistic initialisation of a dual Extended Kalman Filter (DEKF) algorithm using Pseudo Random Binary Sequence (PRBS) battery excitation. A Second-order electrical equivalent-circuit battery model is used whose parameters are identified using a carefully designed 10-bit 10-Hz PRBS signal while the battery is in a zero- or low-current quiescent mode. The PRBS-identified resistive elements of the battery model are then utilised to provide an initial estimate for the battery's SOP. Once in load conditions, the DEKF algorithm is implemented recursively to provide an accurate estimate of the battery's parameters, SOC and subsequently its SOP in real time. The experimental results obtained form an electrochemical impedance spectroscopy (EIS) method give confidence to the performance of the proposed hybrid battery parametrisation technique.