“In response to the engineering demand for precise identification and warning of wind turbine blade icing risk, this study systematically analyzes the meteorological causes of wind turbine icing, forms a risk assessment technology solution that can be applied in engineering, introduces its research progress in wind power disaster prevention, establishes an interpretable machine learning icing decision model based on XGBoost, explores the nonlinear coupling effect of multiple meteorological factors, and provides a solution for scientific planning and precise icing warning of wind farms.”