Adaptive Model Predictive Control for a Novel Hybrid Energy Storage

This paper proposes an adaptive model predictive control for a novel hybrid energy storage(HESS). The proposed hybrid energy storage including battery and supercapacitor is based on modular multilevel converter. The mathematical model for the HESS is established and discrete predictive model is concluded. The cost function for the arm level selection is based on the reference current tracking, and the cost function for the selection of inserted submodules is designed for energy balance among battery units. To respond to parameter mismatch, a parameter estimation method for the proposed HESS based on recursive least square is designed. The estimated parameter value is updated in each control period. The proposed control method is verified by simulation, and is proved to have good steady state performance and fast response speed.