State of charge and model parameters estimation of liquid metal batteries based on adaptive unscented Kalman filter

Abstract Liquid metal battery (LMB) is a newly developing energy storage technology with promising prospect for electric energy storage applications. However, few research efforts have been put into estimating the state of charge (SOC) of LMBs yet, which is of great significance for the battery utilization. In this paper, an adaptive unscented Kalman filter is developed for online estimating the state of a battery model for LMBs, including the SOC and the model parameters. Simultaneously estimated parameters ensure high fidelity of the model and hence facilitate a trustworthy SOC estimation. The algorithm is robust to external disturbances on account of the adaptive estimation of the process and measurement noise covariances. Experimental results show that the proposed algorithm has a satisfactory performance in the estimation of both the SOC and the model parameters.