Unified cost function model predictive control for a three-stage smart transformer

The massive integration of power electronics-based renewable energy sources has profoundly changed the electrical grid. In this scenario, the smart transformer, which is a solid-state transformer with advanced control and communication features, has been proposed as one of the solutions to offer new grid services, while mitigating electrical grid issues, including voltage/frequency disturbance, harmonics, voltage instability, and to pave the way towards dc grids. The commonly proposed topology for ST is the three-stage ac-dc-dc-ac converter, due to the availability of the dc link at both medium- and low-voltage sides. The control design usually relies on the well-known techniques of pole/zero placement and each conversion stage is considered separately. This paper proposes a unified predictive control of the three stages of the ST that allows to control all the variables with a single cost function. Simulation results show the effectiveness of the proposed solution in guaranteeing excellent current tracking performance and good disturbance rejection.