An adaptive nested dynamic downscaling strategy of wind-field for real-time risk forecast of power transmission systems during tropical cyclones

Abstract

High winds from tropical cyclones can cause significant damages to power transmission system and lead to widespread power outages resulting in tangible socio-economic losses. Meteorological numerical weather prediction (NWP) models can provide real-time wind-field forecasts, making it possible to conduct real-time risk forecasts of regional-scale power transmission systems in order to inform proactive risk mitigation interventions. However, the operationally employed NWP simulates wind-fields at mesoscales (1∼10km) which is too coarse to capture the wind dynamics of transmission towers (generally covering an area of 25m*25m or less) and conductors over complex terrains. While dynamic downscaling can resolve topographic features into mesoscale wind-field to obtain small-scale (100m∼1km) or micro-scale (10∼100m) wind predictions, it requires practically unattainable computational resources for regional-scale operations in real-time. To address these challenges, we propose an Adaptive Nested Dynamic Downscaling (ANDD) strategy, which (i) spatially considers terrain features, and topology and failure mechanisms of transmission systems; and (ii) temporally adapts to the evolution of an approaching cyclone in real time, enabling the downscaling domains to be reconfigured based on the latest mesoscale NWP. The advantages of the ANDD strategy are illustrated through the power transmission system in Zhejiang Province (105,500 km2), China, during Super Typhoon Lekima of 2019.