The flow field in the rotor-stator rim cavity is controlled by both the instantaneous distortion caused by rotor and stator airfoils. Naturally one would then assume that the design of optimal rim seal geometries would require full unsteady turbine stage simulations, that would then require an experimental assessment. This manuscript presents a reduced-order approach, neglecting the obvious unsteady interaction effects, to optimize the cavity geometry with a two-dimensional axisymmetric approach, making it possible to explore a wide range of geometries followed by an experimental assessment in a linear cascade. Two parameterization strategies are presented, coupled with a genetic optimizer to maximize thermal protection to the rotor rear platform while minimizing the cooling massflow required. This reduced-order optimization scheme was then assessed at different levels of sophistication to assess the effect of rotation, the vanes, and the rotor geometry with a series of increasing levels of fidelity to the real turbine conditions. Finally, this paper demonstrated the sensitivity of the optimal geometries to variations in the axial gap and purge coolant total pressure.

中文翻译：

---

轮缘密封腔的空气热优化，以增强转子平台的热保护

转子-定子轮缘腔中的流场由转子和定子机翼引起的瞬时变形控制。然后自然会认为，最佳轮辋密封几何形状的设计将需要完全不稳定的涡轮级仿真，然后需要进行实验评估。该手稿提出了一种简化的方法，忽略了明显的非稳态相互作用效应，使用二维轴对称方法优化腔体几何形状，从而有可能探索各种几何形状，然后在线性级联中进行实验评估。提出了两种参数化策略，并结合了遗传优化器以最大化对转子后平台的热保护，同时将所需的冷却质量流降至最低。然后，在不同的复杂度下评估此降阶优化方案，以评估旋转，叶片和转子几何形状的影响，并增加一系列对真实涡轮机条件的保真度。最后，本文证明了最佳几何形状对轴向间隙变化和吹扫冷却剂总压力的敏感性。