Transient 1D Loss Model for Predicting Supersonic Turbine Starting

Abstract Hydrogen rotating detonation engines have demonstrated potential to decarbonize the transportation and power sectors through a step change of engine efficiency improvement. However, the unsteady, low-supersonic flow conditions produced by the combustor pose significant aerothermal challenges for downstream turbomachinery. This study presents a 1D transient loss model for enhanced analysis and loss prediction in axial supersonic inlet turbines subject to highly transient inflow. The supersonic turbine geometry analyzed is optimized for Mach 2 inflow. A 1D loss model accounts for viscous and shock losses, using an equivalent friction factor constructed from the 2D shock structure and viscous correlations. To validate its predictive capabilities, the 1D model is compared against mass flow-averaged results from 2D Unsteady Reynolds-Averaged Navier-Stokes simulations for five inlet Mach numbers ranging from 1.5 to 2.4. Results demonstrate that the 1D model can predict the steady total pressure loss distribution within 3% of the averaged 2D data. In terms of unsteady pressure loss prediction, the 1D model with losses is compared to an isentropic 1D model to reveal the impact of losses during the transient starting process. The 1D model with loss consideration demonstrated an important improvement in predicting transient total pressure losses at the turbine exit with a prediction error reduction of 40 percentage points relative to 2D URANS data, applying a novel loss coefficient for the starting transient. This research provides a computationally efficient, reduced-order tool that can aid in future optimization and analysis of compact, high-work extracting supersonic turbines for rotating detonation engines.