Analysing power system cascading failures and service disruptions in a data-scarce environmentYe, Mengqi (Avtor) Vanniya Perumal, Surender Raj (Avtor) Pantoš, Miloš (Avtor) Ribnikar Cimerman, Sebastijan (Avtor) Ward, Philip J (Avtor) Koks, Elco E. (Avtor) cascading failuresservice disruptionsnetwork analysispower flow modeltropical cyclonesCascading failures in power grids can escalate localised disruptions into extensive service outages. This study presents a transparent, reproducible framework for constructing large-scale power grid models from publicly available data. To assess service impacts on industrial users, we combine industrial points of interest with OpenStreetMap industrial zones and link them to substations using a capacity- and distance-informed allocation method. We examine two disruption scenarios: 1) random failures using percolation analysis and 2) hazard-driven failures using Monte Carlo sampling of wind-induced fragility curves derived from the 2024 Typhoon Yagi wind field. Results show that the Vietnam high-voltage grid is structurally vulnerable, with the giant component fragmenting rapidly under low levels of random bus removal, indicating limited redundancy. Although more than 95 % of transmission lines show no overload under random disruptions, critical corridors around major urban centres exhibit high overload probabilities exceeding 0.5. Under Typhoon Yagi, both direct and indirect failures contribute substantially to disruptions, with indirect impacts extending beyond wind exposed areas. Despite an average served load ratio of 0.91, 26 % of loads lose service in at least one simulation, revealing pronounced spatial disparities in resilience. These findings provide actionable insights for risk-informed resilience planning and targeted reinforcement investment.20262026-03-30 10:03:35Članek v reviji181271sl