High-Altitude Platform Stations or High-Altitude Pseudo-Satellites (HAPS) use propulsion systems which are commonly based on propellers. In this paper, an algorithm for the design of those propellers considering uncertainties is developed and applied. The algorithm is based on the non intrusive polynomial chaos expansion scheme, which converts the stochastic design problem into an equivalent deterministic one. Two uncertainties are studied and characterized: 1) the stratospheric wind fluctuations using reanalysis datasets and 2) the variability of the aerodynamic coefficients caused by the low Reynolds number. The results of the method are analyzed to tackle how relevant the uncertainties are in the propulsion of the stratospheric platforms. The case of study is an ideal stratospheric airship that operates at a mean wind speed of 9 m/s and requires a thrust of 100 N, both uncertain magnitudes. The propeller is built on NACA4412 airfoils and the cost function to be maximized is the mean net propulsion efficiency. The new method provides a relevant gain in the mean efficiency when compared with the deterministic optimization.

高空平台站或高空伪卫星（HAPS）使用通常基于螺旋桨的推进系统。在本文中，开发并应用了一种考虑不确定性的螺旋桨设计算法。该算法基于非侵入式多项式混沌扩展方案，该方案将随机设计问题转换为等效的确定性问题。研究和表征了两个不确定性：1）使用重新分析数据集的平流层风波动； 2）由低雷诺数引起的空气动力学系数的可变性。分析该方法的结果以解决平流层平台推进中不确定性的相关性。研究的案例是理想的平流层飞艇，其平均风速为9 m / s，并且需要100 N的推力，这两个幅度都不确定。螺旋桨建立在NACA4412机翼上，要最大化的成本函数是平均净推进效率。与确定性优化相比，新方法在平均效率方面提供了重要的收益。