Stability analysis and active stabilization of DC microgrid clusters using adaptive controllers

Abstract

This report presents a small-signal model of a DC Microgrid (MG) cluster and its control system. In the closed loop control system, proportional-integral (PI) controllers and two active stabilization methods, namely the droop loop and the feed-forward compensation loop, are used. A consensus-based distributed control framework is used to control the output voltages of the MGs and power flow between the MGs. Simulations of the small-signal model on MATLAB/Simulink demonstrate that the control system successfully stabilizes the DC MG cluster at a particular operating point, as it is able to reject sudden changes in voltage, current, or load demand. The PI controllers are then replaced with adaptive PI controllers, and simulation results show that the adaptive PI controllers better stabilize the system against sudden load changes and current disturbances. Hardware-in-the-loop simulations on OPAL-RT verify that the adaptive PI controllers and feed-forward compensation loop stabilize the system more rapidly, as the MG voltages settle faster to the steady state. This report concludes with some possible future extensions to this project, and my reflections.