Aero-thermal performance validation of topology optimized gas turbine guide vane : CFD, CHT analysis and validation of a Topology Optimized guide vane using RANS

Sammanfattning

Gas turbines operate under extreme thermal conditions, making efficient cooling systems essential for maintaining performance, reliability, and the longevity of critical components. This thesis investigates the aero-thermal performance of topology-optimized cooling solutions for gas turbine components, validated through advanced Computational Fluid Dynamics (CFD) simulations. Topology Optimization (TO) was employed to create innovative cooling designs. These TO designs were initially created using Reynolds-Averaged Navier-Stokes(RANS) with a zero equation turbulence model, providing a preliminary evaluation of flow characteristics and heat transfer performance. To ensure the practical viability of these designs under realistic operating conditions, they were subsequently validated through RANS simulations using the k-omega SST turbulence model, which effectively captures near-wall behavior and boundary layer interactions. Following this, further refinement was conducted using the k-epsilon Lag Elliptic Blending turbulence model to enhance predictive accuracy in complex flow regions, especially in areas with flow separation and recirculation, where traditional models may fall short. A grid convergence index (GCI) study was also performed to quantify the numerical uncertainty associated with the simulations, thereby ensuring the robustness and reliability of the CFD analyses. The results demonstrate that TO-generated cooling designs significantly improve cooling efficiency, offering a higher level of control over flow separation and thermal performance in high-temperature environments. The integration of CFD models allowed for an accurate representation of flow and thermal behaviors in ribbed channels, supporting the broader application of TO methods in gas turbine cooling systems. This research contributes to the development of more effective and efficient cooling solutions and outlines potential areas for future improvement in TO methodologies and turbulence modeling for practical applications.