A Master-Slave Model Predictive Control Approach for Microgrids

This article proposes a master-slave finite control set model predictive control (FCS-MPC) for microgrids. To demonstrate it, a microgrid is considered, composed of a master neutral-point clamped (NPC) inverter with a battery energy storage system (BESS) and output <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">LC</i> filter; two slave NPC inverters with photovoltaic (PV) panels and output <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">LCL</i> filters; RL and nonlinear loads. Two modes of operation are proposed for the primary control of the microgrid. In the first, the microgrid is connected to the main grid, and the master and slaves are grid-following inverters. In the second, the microgrid is islanded, and the master is a grid-forming inverter, while the slaves remain as grid-following inverters. To validate the performance of the proposed master-slave FCS-MPC, hardware-in-the-loop (HIL) results are presented for different operational conditions of the microgrid, including grid connection, transition to islanded mode, and load variations. The results demonstrate the good performance of the proposed master-slave FCS-MPC, such as fast dynamic response, multivariable control, and robustness to parametric uncertainties and variations.