A Novel Model-Based Estimation Scheme for Battery-Double-Layer Capacitor Hybrid Energy Storage Systems

Battery-double-layer capacitor (DLC) units are becoming popular hybrid energy storage systems (HESS) for vehicle propulsion, auxiliary power units, and renewable energy applications. Safe and optimal operation of the HESS requires real-time monitoring of its constituent subsystems. In this paper, we use a model-based approach to monitor HESS behavior and propose an online combined state and parameter estimation scheme using coupled electrical-thermal dynamical models for each subsystem. In particular, we focus on an HESS composed of a lead-acid (PbA) battery and a DLC for which experiments have been designed to identify the initial model parameters. The estimation scheme uses a novel cascaded observer-based structure which: 1) is designed based on sliding mode methodology and 2) exploits the coupling of the electrical and thermal dynamics. Using Lyapunov's arguments, theoretical conditions are derived which characterize the convergence of the state and parameter estimates in the presence of uncertainties. The effectiveness of the estimation scheme is evaluated via simulation and experimental studies on the PbA battery, the DLC, and the HESS system.