Data‐driven short‐circuit detection and location in microgrids using micro‐synchrophasors

Fault detection and location is a challenging issue in microgrid protection, which is increasingly more complex in the presence of distributed generators based on renewable energy, due to their inherent intermittency. In this context, a novel data-driven approach for fault detection and location in microgrids is proposed, by using graph theory representation and micro-synchrophasors also known as μ PMUs. This proposal adopts the conviction to provide an accurate fault location even under variations in short-circuit levels caused by the intermittency of distributed generators. This is in sharp contrast with traditional short-circuit rating-based methods, which are not always advisable due to the intermittent nature of power sources. This work proposes the use of a modelling specification in terms of equilibrium equations, that can reveal not only the underlying physical laws of the netowork, but also the occurrence and location of short circuits based on phasor data. The intermittency of distributed generation is modelled in the proposed approach, which permits to yield trustworthy information either distributed generators are involved in a fault or not. As a consequence, the fault location errors are significantly reduced during the fault location process. The theoretical findings of this proposal are validated via simulation results and experiments using commercial micro-synchrophasors and hardware-in-the-loop emulation of a realistic microgrid.