Evaluation of Empirical Loss Models for Base Drag Prediction of sCO2 Axial Turbine Aerofoils

Abstract

Abstract Axial flow turbines can be efficiently designed and used in megawatt scale sCO2 closed Brayton power systems. Empirical loss models are required for design and optimization of axial turbines. Ample loss models are available in the open literature for air and steam based axial turbines. But no loss models have been reported so far exclusively for sCO2 applications. Constructing a complete loss model for sCO2 turbomachinery is farfetched until loads of data are generated from cascade tunnels and turbomachinery test rigs with sCO2 as the operating fluid medium. Unlike air and steam, which are approximated as ideal gases, sCO2 is a real gas. Nevertheless, designers and commercial turbomachinery design softwares use corrected versions of these existing loss models for sCO2 turbine design. The profile loss / base drag contributes to 50–60 % of overall losses in an axial turbine. Estimation of base profile drag is important at the beginning of the design phase of the turbine in order to maximize the power output and efficiency. An attempt has been made in this paper to evaluate some of the well proven axial turbine loss models like Ainley & Mathieson, Dunham & Came, Kacker & Ockappu, Aungier etc. for estimating base profile drag of an sCO2 axial turbine aerofoil. The aerofoil investigated in the present study has been designed for a 10 MW sCO2 axial turbine stator. Profile drag of the aerofoil was computed from full turbine 3D CFD and 3D cascade CFD simulations with sCO2 fluid medium. Turbine CFD simulations were carried out using pressure based ANSYS CFX RANS solver with SST turbulence model. 3D cascade CFD simulations were carried out using pressure based ANSYS Fluent RANS solver with SST k-ω turbulence model. sCO2 fluid properties were modelled using Refrigerant Gas Property tables generated from NIST Refprop database. It would of great interest to see if cascade wind tunnel data generated with air can be applied to sCO2 turbine design. In this context, cascade test experiments were carried out for a 10 bladed linear cascade model of the aerofoil in a high-speed cascade wind tunnel at CSIR-NAL. The aerofoil performance was investigated at the design condition and over a range of off-design conditions. The base profile drag data obtained using the loss models, CFD simulations and cascade test experiments are compared and discussed in detail. This study shows that loss models developed for air-based turbines can be efficiently used to predict base profile drag of sCO2 turbines with corrections. It is also inferred that the cascade wind tunnel data can be a useful tool to study sCO2 aerofoils.