Quantification of Performance Uncertainty for a Transonic Compressor Rotor Using an Adaptive NIPC Method

Abstract

Abstract Turbomachines are designed and run at specific operational conditions. However, due to some deterministic and stochastic sources of flow variations, especially at the inlet and outlet of turbomachinery blades, the influence of operational condition variations to aerodynamic performance are considerable. The quantification of performance change is one crucial module in robust design of turbomachinery blades. The paper studies the performance impact of stochastic operational condition variations using polynomial chaos. To reduce the computational cost, an adaptive sparse grid technique is employed to construct the adaptive non-intrusive polynomial chaos (NIPC) model. Through statistical evaluations of performance change for a three-dimensional transonic compressor rotor blade, NASA Rotor 67, the response performance of adaptive NIPC is firstly verified by comparing with the direct MCS method and static NIPC. Then the adaptive NIPC is used to statistically evaluate the adiabatic efficiency change of Rotor 67 due to the separate variation of inlet total pressure and outlet back pressure at two operation conditions: near peak efficiency and near stall. It is found that the performance change is nonlinear dependent on the inlet and outlet pressure variations. Besides, the simultaneous effects of inlet and outlet pressure variations to the performance change are investigated. Finally, the statistical results of flow variations along span, on the suction side and blade-to-blade streamsurfaces at different spans are illustrated in detail to demonstrate how the inlet and outlet pressure variations influence the aerodynamic performance change.