Typhoon Warning Technology for Offshore Wind Farms Based on Numerical Forecasting

Abstract Typhoons are the most significant hazardous weather events affecting the safe operation of coastal wind farms in China. Due to the scarcity and high cost of observational data, numerical simulation has become a crucial approach for fine-scale typhoon forecasting. Taking the super typhoon Doksuri as a case study, this paper employs the WRF mesoscale numerical model to simulate its track and boundary-layer wind characteristics, and establishes a typhoon early warning system for wind farms based on the simulation results. The results show that before and after landfall, hub-height wind speeds exhibit a pronounced spatial asymmetry, with wind speeds exceeding 50 m/s in the eyewall region and a low-wind zone near the typhoon center. The wind direction may abruptly change by more than 60° within 10 minutes, and after landfall, the area of strong directional variability shifts to the left side of the typhoon's movement path. The near-surface wind shear in the boundary layer differs significantly between sea and land, generally exceeding 0.11 over land. The wind attack angle exhibits strong alternating positive and negative variations with height. Based on these features, this study proposes a multi-indicator early warning framework, including extreme wind-speed warning, rapid-yaw warning, and comprehensive risk warning. On this basis, an integrated forecasting–warning–response typhoon early warning system for wind farms is developed. The results demonstrate that the system effectively identifies key risks and provides a scientific basis for anti-typhoon measures and the rational utilization of typhoon wind energy.