Multi-State Reliability Modeling and Evaluation of Islanded Microgrids with Semi-Active Hybrid Energy Storage Systems

Accurate reliability assessment of islanded microgrids (iMGs) increasingly depends on intelligent system modeling and data-driven integration between physical assets and digital control frameworks. To address the limited adaptability of conventional reliability models, this paper proposes an ICT-integrated framework for reliability evaluation of iMGs equipped with semi-active lithium-ion battery-supercapacitor hybrid energy storage systems (HESS). The framework employs a 32-state Markov-based model to represent component-level degradation and failure dynamics, enabling state-aware reliability analytics through automatic state aggregation and transition probability learning. These states are further abstracted into representative operating modes that can be seamlessly interfaced with energy management systems for online evaluation. A minimum load curtailment strategy is embedded within a time-series Monte Carlo simulation environment to quantify the interactive impact between multi-state HESS behavior and overall system reliability. Standard indices-outage frequency, duration, and availability-are computed to characterize resilience under varying storage conditions. Comparative results verify that the proposed model substantially improves the reliability of iMGs by enabling degraded yet continuous operation during partial failures. The study provides a scalable and digitally implementable reliability framework for HESS-enabled microgrids, bridging the gap between detailed component modeling and real-time operational analytics.