Abstract:The present study proposes a simple grey-box identification approach to model a real dc-dc buck converter operating in continuous mode using a nonlinear polynomial autoregressive with exogenous input (NARX) model. The proposed approach casts the grey-box identification problem into a multi-objective framework and explicitly uses a priori information about the static nonlinearity into the structure selection process. The multi-objective framework could identify globally valid models with improved dynamic prediction and can mimic the static behavior of the buck converter over a wide input range.