Optimization of wind turbine blade airfoil considering environmental parameters uncertainty under rime conditions

In cold climates, ice accumulation on wind turbine blades degrades aerodynamic performance. Airfoil optimization can improve aerodynamic shapes, but uncertainty in environmental parameters poses challenges. This study quantifies the uncertainty of four parameters, including liquid water content, temperature, relative wind speed, and droplet median volume diameter, by virtue of the Latin hypercube sampling method. The NACA 4418 airfoil is optimized using the Non-dominated Sorting Genetic Algorithm II (NSGA-II) and the Multi-Objective Evolutionary Algorithm based on Decomposition (MOEA/D), incorporating uncertainty quantification. Results show that liquid water content has the most significant impact on aerodynamic performance, while relative wind speed has the least effect. The optimized airfoil set shows a 31% average increase in lift-to-drag ratio after icing (C L /C D , ice ), with key design changes such as a raised upper surface at the leading edge, reduced maximum thickness, and a forward shift in thickness location. The optimized airfoil improves performance both before and after icing and ensures reduced aerodynamic performance variability under various icing conditions. After icing, the maximum lift coefficient (C L , ice ) increases by 17.7%, the C L /C D , ice improves by 29.2%, and the trailing-edge separation bubble is eliminated. This method contributes to an effective solution for wind turbine blade de-icing design that balances improved aerodynamic performance with environmental adaptability.