State-of-charge estimation of rechargeable battery with hysteresis characteristics using robust gain-scheduled observer

In this paper, we consider model-based state-of-charge (SOC) estimation of a rechargeable battery with hysteresis characteristics. A standard approach is to describe the battery system as an approximated linear time-invariant model and to apply a linear estimator such as the robust observer or Kalman filter. However, this cannot achieve sufficient estimation accuracy due to the nonlinearity of the hysteresis characteristics. We propose to describe the battery system as a linear-parameter-varying (LPV) model, which does not require the approximation. Furthermore, parameter uncertainties are taken into account to derive robust gain-scheduled observer design. The effectiveness of the proposed method is illustrated through numerical experiments.