Robust optimization design of single-stage transonic axial compressor considering the manufacturing uncertainties

Manufacturing uncertainties always lead to significant variability in compressor performance. In this work, the tip clearance uncertainties inherent in a transonic axial compressor are quantified to determine their effect on performance. The validated tip clearance losses model in conjunction with the 3D reynolds averaged navier-stokes (RANS) solver are utilized to simulate these uncertainties and quantify their effect on the adiabatic efficiency, total pressure ratio and choked mass flow. The sensitivity analysis method is employed to figure out which parameters play the most significant roles in determining the overall performance of compressor. To propagate these uncertainty factors, the non-intrusive polynomial chaos expansion (PCE) algorithm is used in this paper and the probability distributions of compressor performance are successfully predicted. A robust design optimization has been carried out based on the combination of the genetic algorithm (GA) and the uncertainty quantification (UQ) method, leading to a robust compressor rotor design for which the overall performance is relatively insensitive to variability in tip clearance without reducing the sources of the manufacturing noise. The optimization results show that the mean value of the adiabatic rotor efficiency is improved by 1.4 points with the overall variation of that reduced by 64.1%, while the total pressure ratio is slightly improved when compared to the prototype.