Partial Admission Loss Modeling of Radial Inflow Turbines

Abstract Radial inflow turbines (RITs) are widely used in aerospace, energy, and chemical industries, and often operate under challenging conditions such as partial admission (PA). However, there is a paucity of fast and accurate empirical models regarding the PA performance prediction of RITs. Therefore, a PA loss model for RIT off-design conditions has been developed in the present study, based on a loss mechanism analysis using numerical simulations. First, the off-design performance of an RIT was simulated using CFD and validated against experimental data. Then, systematic simulations were conducted to analyze the dominant loss mechanisms under PA conditions, namely windage and pumping losses. Inspired by the PA loss modeling approaches used for axial flow turbines (AFTs), a PA loss model for RITs was developed. Correction coefficients were calibrated using CFD data, accounting for key physical parameters including the partial admission ratio, rotational speed, outlet static pressure, and spatial distribution or number of admission segments. The developed model was applied to several RIT cases and validated numerically under varying PA conditions. The results show an average discrepancy of 1.071% in isentropic efficiency between the model predictions and CFD data. The validated PA model exhibits enhanced accuracy and generalizability, providing an effective tool for preliminary design and analysis of RITs at PA conditions.