Robust design of aero engine structures: Transferring form error data when mapping out design spaces for new turbine components

Paper i proceeding, 2016

In aerospace modeling and simulation, nominal geometries are norm. However, it has been shown that form error, or irregular deviations in geometry, aggravates thermal stresses, which in turn reduces product life. While form error can be measured on manufactured products using 3D laser scanners, a simulation infrastructure is needed to analyze its effects on aerodynamic, structural and thermal performance. Moreover, in early product development phases, before manufacturing has begun, form error data is not available. This paper describes a method for including form error data in mainstream simulation activities. The suggested method works by creating parametric CAD-models to accommodate form error. There are two main benefits of this method. Firstly, it enables proactive robustness simulations where substantial design changes can be tested and evaluated. Secondly, it enables the mapping of data from previous products onto new designs, which means that robustness analyses can be performed in earlier design phases. To demonstrate this capability, a case study shows how a robust optimization scheme using genetic algorithms can improve product robustness to form error. The results show that form error have effects of the same order of magnitude as key design parameter changes. This finding underlines the importance of performing form error analyses in exploratory early design phases.