Adaptive controllers for grid-connected DC microgrids

Abstract

Droop technique is extensively adopted in dc microgrids, for both islanded and grid-connected modes of operation. However, voltage regulation and proper current sharing are considered as the major challenges in droop techniques. Moreover, changes in system parameters and wide load demand variations degrade the operational performance and the efficiency of the dc microgrid. In this paper, a finite-state-machine (FSM) based controller for the inner loop of microgrid converters is utilized. Although the FSM controller operates in voltage mode and requires no current sensors, the dynamic performance of the microgrid converters is limited. Thus, a distributed consensus output voltage tracking based on the performance recovery controller is presented which adopts nonlinear observers for the microgrid converters to account for model uncertainties and disturbances due to parameters variations. The dynamic performance and robustness of the FSM controller and performance recovery controller are assessed using PSCAD/EMTDC simulation. Moreover, a comparative study with the classical Proportional-Integral (PI) controller is presented, considering both rating-based and cost-based current sharing schemes of the dc microgrids.