The invention discloses a small wind turbine wing type aerodynamic robust optimization design method suitable for turbulent working conditions, belonging to the technical field of wind turbines. The invention adopts the non-embedded probability collocation point method in the design to realize the quantitative characterization of the turbulence intensity of the uncertain parameter. Through the improved Hick-Henne type function parameterization method, Latin hypercube test method and CFD numerical calculation method, the kriging agent model between design variables, uncertain parameters and themaximum lift-to-drag ratio of the airfoil is established. On this basis, a robust optimization mathematical model is established to maximize the mean value of the maximum lift-to-drag ratio and minimize the standard deviation of the wind turbine wing under uncertain turbulent conditions. The non-embedded probabilistic configuration point method, the kriging proxy model, and the NSGA-II optimization algorithm are used to optimize the wind turbine airfoil. The invention improves the maximum lift-drag ratio of the airfoil, reduces the fluctuation range, improves the wind energy capture efficiency, and enhances the aerodynamic robustness under the turbulent condition. At the same time, the computational workload of robust optimization design is reduced, and the optimization efficiency is improved, which provides an important reference for the optimization design of wind turbine wing under turbulent conditions.