Antar Kumar Biswas

Masoud H. Nazari, Hamid Varmazyari, Antar Kumar Biswas et al.

This paper presents a comprehensive review of community-centric power system resilience, emphasizing the integration of community-level resilience considerations and techno-legal governance frameworks with engineering-based resilience enhancement strategies and data-driven approaches to address extreme events. Recent large-scale outages have demonstrated that power disruptions can cascade beyond electrical infrastructure and disproportionately affect vulnerable communities, critical services, and interconnected urban systems, highlighting the need for resilience approaches that integrate technical, social, and regulatory dimensions. Within this community-centric perspective, the review first summarizes state-of-the-art strategies for enhancing power system resilience, including network hardening, resource allocation, optimal scheduling, and system reconfiguration techniques, while highlighting the growing role of artificial intelligence (AI) and data-driven analytics in supporting resilience planning and operational decision-making. It then examines the interdependencies between power system resilience and community resilience, addressing socioeconomic and behavioral dimensions, cross-infrastructure interconnections, and the emerging role of resilience hubs. The paper further examines the techno-legal frameworks governing resilient energy systems by comparing the regulatory landscapes of the European Union (EU) and the United States, highlighting key similarities and distinctions that shape resilience planning and implementation. By analyzing state-of-the-art engineering-based, AI-driven, and techno-legal methods for assessing and mitigating the impacts of high-impact, low-probability (HILP) events, the review identifies critical research gaps and outlines promising directions for future investigation.

Nina Fatehi, Antar Kumar Biswas, Masoud H. Nazari

This paper presents a novel learning based framework for predicting power outages caused by extreme events. The proposed approach targets low-probability high-consequence outage scenarios and leverages a comprehensive set of features derived from publicly available data sources. We integrate EAGLE-I outage records from 2014 to 2024 with weather, socioeconomic, infrastructure, and seasonal event data. Incorporating social and demographic indicators reveals patterns of community vulnerability and improves understanding of outage risk during extreme conditions. Four machine learning models are evaluated, including Random Forest (RF), Graph Neural Network (GNN), Adaptive Boosting (AdaBoost), and Long Short-Term Memory (LSTM). Experimental validation is performed on a large-scale dataset covering counties in the lower peninsula of Michigan. Among all models tested, the LSTM network achieves higher accuracy.