A Model-Predictive-Control Strategy for Hybrid Dual Active Bridges Based DC Transformer

In order to improve dynamic performance and robustness, a model predictive control (MPC) strategy is proposed for a hybrid dual active bridges (HyDABs) based DC transformer (DCT). Based on the power transmission characteristics, a dynamic first-order linear discretization model is established. Moreover, a two-step-ahead prediction algorithm is used to deal with one-sampling time delay. The traditional voltage closed-loop control (TVLC) scheme with load current feedforward and the proposed MPC scheme are compared on a downscale four-cell laboratory prototype. In addition, the PowerPC P2020 + FPGA_ 6SLX45 is used as the controller. The excellent dynamic performance with MPC algorithm is validated by extensive experimental results as suffering voltage fluctuation and power step.