A Critical Aspect of Dynamic Stability in Autonomous Microgrids: Interaction of Droop Controllers Through the Power Network
Mohsen Eskandari, Andrey V. Savkin
Abstract
It is explored in this article that the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">interaction of droop controllers through the power network</i> (IDCPN) is the dominant factor affecting the dynamic stability of an autonomous networked microgrid (ANMG) and new models are developed to support the IDCPN. The models are developed to analyze the impacts of three parts on the IDCPN: 1) the X/R ratio of the power network impedance; 2) the droop controllers including low-pass filters, and 3) the impedance characteristics of the grid-forming voltage source inverters (VSIs). The low-frequency oscillations (LFO), excited by IDCPN, is identified and quantified by modeling the first and second parts. The critical impact of the low X/R ratio on amplifying the LFO is clarified. Then, the output impedance of the grid-forming VSI, including the virtual inductance loop (VIL), is developed and rigorously observed that reveals inefficient performance. It is shown that the VIL improves dynamic performance by providing sufficient damping to suppress LFOs and not by effectively boosting the X/R of the VSI output impedance. This, however, is problematic in the current limiting that puts the ANMG at instability risk because of the resistive–capacitive impedance characteristics of the VSIs. A <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">${{\boldsymbol{H}}_\infty }$</tex-math></inline-formula> robust controller is proposed to replace VIL for suppressing LFO and stabilizing ANMG. Numerical/simulation results are provided to prove the accuracy of the models.