High Dynamic Current Control Strategy for Bidirectional Isolated AC-DC Matrix Converter Based on Deadbeat Predictive Algorithm

In the conventional double -closed-loop control strategy for the bidirectional isolated AC-DC matrix converter, many control parameters are overmany and it is difficult to be tuned. In order to solve these problems, a high dynamic current control strategy based on deadbeat predictive algorithm is proposed in this paper. In the proposed strategy, a deadbeat predictive algorithm is introduced for the grid currents control, by which the controller parameters are reduced and digital control delay is eliminated. A large number of voltage and current detection in the algorithm are taken into account, and the input voltage observer is used to reduce the detection cost and the impact of sampling errors. The transducer uses a single phase shift method to coordinate with the deadbeat predictive algorithm. The reference input current and the phase shift angle are bijective functions, and the method is simple and easy to implement. Simulation and experimental results show that the proposed control strategy can achieve the sinusoidal grid currents and a unity power factor. The DC voltage is stable with a current ripple rate of less than 1%. Fast-tracking is realized with no overshoot and no oscillation when the DC current changes abruptly at the reference value. Meanwhile, the proposed control strategy reduces the current dynamic adjustment time by 69%, 85% and 67%, respectively, compared with the conventional double-closed-loop control strategy during the forward and reverse power transfer and forward and reverse switching of the converter. It can be verified that the proposed control strategy can ensure good input and output performance of the converter and significantly improve the dynamic performance in the switching process compared with the conventional double-closed-loop control strategy.