Efficient surrogate-based robust shape optimization for vane clusters

The paper is aimed to describe the robust optimization process for a shape optimization of a low pressure turbine (LPT) guide vane cluster. The multidisciplinary analysis, needed to evaluate multidisciplinary constraints, which drive the vane cluster design, is described. Attention is paid to the different sources of uncertainty, such as manufacturing tolerances, material parameters and performance parameters. An overview of the recent methods for the Reliability-based Robust Design Optimization (RBRDO) is presented, with the focus on efficient adaptive surrogate-based method. Finally, a new surrogate-based approach is introduced, which is inspired by the Efficient Global Optimization (EGO) method1 and inverse reliability analysis.