The aero-thermodynamic design of supercritical CO2 radial turbine based on the particle swarm optimization and vortex competitive mechanism

The supercritical carbon dioxide (S-CO2) radial inflow turbine (RIT) is a critical component in advanced power cycles. However, its design is challenged by the complexity of thermophysical properties of S-CO2 and the interdependent nature of key empirical parameters. This study establishes a robust one-dimensional aero-thermodynamic design methodology for S-CO2 RITs, integrating an optimized loss model correlation with the Particle Swarm Optimization (PSO) algorithm. The PSO algorithm was then employed to automate the synergistic optimization of five critical dimensionless parameters: reaction degree, flow coefficient, velocity ratio, incidence angle, and radius ratio. The application of this framework to a 350 kW case study demonstrated a significant performance enhancement, achieving a 1.67% increase in total-static efficiency and a 2.03% gain in output power compared to a baseline design. Flow field analysis revealed that the optimized design, characterized by a higher reaction degree and increased blade height, effectively suppresses tip leakage flow and mitigates the adverse coupling between tip leakage vortices (TLVs) and secondary flows by leveraging controlled vortex interactions. This mechanism fundamentally reduces passage and clearance losses, thereby validating the proposed multi-parameter optimization approach as a powerful tool for the high-performance design of S-CO2 RITs.