Improved Alternating Sequential Model Predictive Voltage Control of Direct Matrix Converter for Medium-Frequency Power Supply

The alternating sequential model predictive control (ASMPC) dynamically equalizes the priorities of multiple control objectives by alternately executing the cost functions of the control objectives. It can balance the control performance of multiple objectives without the weighting factor. However, it is easy to miss the optimal switching state for equalizing multiple control objectives when alternately executing the cost functions of the control objectives. In this paper, an improved alternating sequential model predictive control (IASMPC) for matrix converters (MC) is proposed, which can maintain the advantages of the ASMPC while improving the way of selecting the cost function of the control objectives and selecting the optimal switching states by dynamically adjusting the number of selected switching states. The proposed control strategy is used in the medium-frequency power supply (MFPS) based on a direct MC, improving the power current control performance while maintaining the output voltage control performance. The simulation results verify the feasibility and effectiveness of the proposed control strategy.