Uncertainty quantification of the inlet boundary conditions in a supercritical CO2 centrifugal compressor based on the non-intrusive polynomial chaos

Given the sensitivity of supercritical carbon dioxide centrifugal compressors performance to inlet boundary parameter fluctuations, this study focuses on the uncertainty quantification of the inlet boundary conditions. Non-intrusive polynomial chaos surrogate models were developed for both performance parameters and flow field of a supercritical carbon dioxide centrifugal compressor. Sensitivity and correlation analyses revealed that inlet total temperature is the primary influencing parameter. Fluctuations of 1.37 % in total temperature and 6.75 % in total pressure result in changes of 40.94 %, 18.57 %, and 1.77 % in mass flow rate, total pressure ratio, and total to total isentropic efficiency, respectively. Flow field statistical analysis shows that inlet uncertainties nonlinearly impact the dryness fraction and relative Mach number distributions near the blade leading edge, while the density in the condensation zone on the suction side of the leading edge is less affected. Error bands in static pressure on blade surfaces highlight significant variations, particularly at 90 % spanwise on the blade suction side, where shocks and supersonic flow induce prominent disturbances. This study offers a preliminary methodology and theoretical guidance for robust aerodynamic optimization and operational strategies of supercritical carbon dioxide centrifugal compressors.