Predictive current control for indirect matrix converter with reduced current ripple

Abstract

The predictive current control (PCC) for an indirect matrix converter (IMC) with reduced current ripple is presented in this study. In the proposed PCC scheme, an IMC drives a load applying a split switching vector, which is used as a candidate vector of the cost function. The cost function of the proposed PCC includes a reference state and a predicted state. With the two states, the optimal output vector can be selected from the split switching vectors, thus alleviating the current ripples on both the grid side and the load side of the IMC. However, the use of the split switching vector increases the computation complexity of the cost function optimization because the number of predicted states is increased by the split switching vectors. Hence, the computation complexity reduction method in a deadbeat fashion is proposed for the simple implementation of the proposed PCC on a digital signal processor. In addition, the unity power factor on the grid side is guaranteed by a straightforward modulation technique when the maximum voltage is transferred. The performance of the proposed PCC for the IMC is verified by simulation and experimental results.