A Dynamic Response Enhanced Model Predictive Control for Bidirectional Dual Active Bridge Converter

Model Predictive Control (MPC) can enhance the dynamic response performance of Dual Active Bridge (DAB) converters, but its strong dependency on model parameters leads to poor system robustness. To address this issue, this paper proposes an ultra-local model based MPC (ULM-MPC) strategy This method establishes a ULM that retains only accurately measurable parameters, categorizes all remaining uncertain parameters as a lumped disturbance, and calculates the lumped disturbance in real-time using parameter identification techniques. Without altering the original control structure, this method eliminates control accuracy degradation caused by model parameter inaccuracies, thereby improving system robustness. Furthermore, building upon dual-phase-shift modulation, the Lagrange multiplier method (LMM) is employed to solve optimal modulation parameters, achieving current stress optimization and enhancing steady-state operational efficiency. Finally, experimental verification is conducted to validate the effectiveness and superiority of the proposed method.