ANN based on MPC for a three-phase active rectifier

The three-phase rectifiers are widely used in different industrial applications, like energy storage systems, uninterruptible power supplies, renewable energy generation systems due to their high-power transfer and its power factor correction capability with high-quality DC output and sinusoidal input currents. To control the power factor, the model predictive control technique has emerged as a suitable and intuitive control strategy, however, the high computational cost limits the converter frequency operation. As a solution, in this paper an artificial neural network based on model predictive control is proposed to obtain a three-phase sinusoidal input currents under different power demands, reducing the computational cost.