Experimental and Numerical Studies of the Aerodynamics of Stationary Two-Shaft Gas Turbine Exhaust System

Abstract

In this study, the aerodynamic performance of the exhaust system of a two-shaft gas turbine was investigated experimentally and numerically. The investigation focused on the system “Turbine Stage-Diffuser—Collector Box” and aimed to examine the impact of inlet conditions and geometry particularities on the efficiency of the exhaust system. The experiments were conducted on the Test Ring ET4 (Experimental Turbine-4) at the Peter the Great St.Petersburg Polytechnic University, which was equipped with a special diversion channel to examine the non-axisymmetric outlet of the exhaust duct. The collector box was designed to rotate by 180 degrees around the turbine axis to investigate its impact on the system’s performance. Flow traversing parameters were measured with the five-channel pneumatic pressure probes, and numerical simulations were performed with CFX 15.0. The RANS (Reynolds-averaged Navier–Stokes) equations were closed with the SST (k-ω) turbulence model (Shear Stress Transport model). The study concluded that the RANS SST model predicts the flow in the diffuser before the struts accurately. However, downstream the struts, the CFD (Computer fluid dynamic) results over-predicted the exhaust diffuser pressure recovery coefficient by 14% due to the complex vortex structure of the turbulent flow, which the Averaged Navier–Stokes equations did not resolve. The study highlights the importance of considering the last stage of the turbine, diffuser, and collector box as an integrated system when investigating the aerodynamics of exhaust ducts. The study also emphasizes the impact of geometry and inlet conditions on the exhaust diffuser’s performance and efficiency. The results of this study can be used to optimize the design of the exhaust system of two-shaft gas turbines and improve their thermal efficiency. The integrated approach of combining experimental and numerical methods can provide a detailed and reliable flow picture and can be used for future research in this area.