Home // UBICOMM 2025, The Nineteenth International Conference on Mobile Ubiquitous Computing, Systems, Services and Technologies // View article

Adaptive Microgrid Architecture to Manage System Resiliency

Authors:
Mobolaji Bello
Davis Montenegro
Oladayo Bello

Keywords: Adaptive control; machine learning; microgrid; robust control; power system resilience.

Abstract:
In light of the growing risks posed by high-impact, low-frequency events (such as those driven by climate change and other emerging hazards) utilities are increasingly deploying Distributed Energy Resources (DER), both utility- and customer-owned, to enhance grid reliability. These assets play a vital role in addressing system constraints during peak demand (thermal and voltage), mitigating power outage impacts, and improving overall resilience by supporting the formation of microgrids when distribution grid integrity is compromised. Yet, microgrid deployment presents its own technical challenges, particularly in coordinating the DERs involved. Critical functions; for example, grid separation (islanding), black start procedures, operational control, and eventual grid reconnection, must be executed with precision to ensure system stability. Poor coordination can exacerbate existing grid disturbances, extend recovery timeframes, and ultimately undermine the very resilience the microgrid is intended to deliver. To address these challenges, this paper proposes a microgrid architecture anchored by three resilience-enhancing pillars: (1) robust protection and power quality, (2) high-speed, reliable communication infrastructure, and (3) Machine Learning (ML) driven control and management. Each pillar is introduced through its operational goals and technical contributions, followed by a test case illustrating the integrated architecture in action.

Pages: 8 to 13

Copyright: Copyright (c) IARIA, 2025

Publication date: September 28, 2025

Published in: conference

ISSN: 2308-4278

ISBN: 978-1-68558-288-3

Location: Lisbon, Portugal

Dates: from September 28, 2025 to October 2, 2025