Robust design optimization considering inlet flow angle variations of a turbine cascade

Abstract

The stochastic variation of inlet flow is one common uncertainty in turbomachines, the performance impact of which requires to be quantified and taken into account in the design optimization of turbomachinery blades. The paper introduces the robust aerodynamic design optimization improving the mean performance and the robustness of a turbine cascade at off-design condition. The consistent-refinement of the grids used for solving the Euler equations is checked and the employed numerical scheme is verified and validated. Performance impact of the inlet flow angle variations is statistically evaluated by using a non-intrusive polynomial chaos with an adaptive sparse grid sampling method. The camber of the turbine blade is redesigned with the enforced blade thickness distribution to minimize the statistical mean and standard deviation (Stdv) of total pressure loss coefficient. The results of both the deterministic and robust aerodynamic design optimization are presented and compared in detail. The effects of camber redesign on reducing the statistical mean and Stdv of total pressure loss coefficient are discussed. Moreover, the statistical flow solutions are illustrated to demonstrate the influence mechanisms of flow variations in the blade passage on the improvements of aerodynamic robustness.