On the similarity and challenges of multiresonant and iterative learning current controllers for grid converters and why the disturbance feedforward matters

摘要

There are two main techniques to solve the reference tracking problem for repetitive references and under repetitive disturbances, namely multiresonant (a.k.a. multioscillatory) controllers and iterative learning controllers. Nevertheless, neither of the approaches is a definitive winner, which is to be demonstrated herein. Both have their strengths, weaknesses and challenges. A grid-tie converter will be the case study here. The goal is to draw or inject sinusoidal currents under distorted grid voltage conditions. The supporting feedforward controller will be addressed within the context of the discussed repetitive control task. The case will be illustrated using numerical simulations. Our main goal is to make practitioners familiar with the relationships between these two control methods.