Multi-Objective Optimization of Aerodynamic Performance for a Small Single-Stage Turbine

Abstract

In order to improve the performance of single-stage turbines, blade profiling optimization was conducted for the guide vane and rotor under design condition. Support vector regression (SVR) and non-dominated sorting genetic algorithm-II (NSGA-II) were used to execute the optimization, with the objective of maximizing the efficiency and total pressure ratio of single-stage turbines. The gas turbine chosen for the initial study was the KJ66, which is one of the most robust and primitive small gas turbine designs available. The influence mechanism of the stator and rotor profiling on flow field and performance was discussed. The results revealed that compared with the prototype, the adiabatic efficiency increased by 5.95% and the total pressure ratio increased by 0.9%. Furthermore, the matching of flows between the stator blade and rotor blade obviously improved. The optimized guide vane suppressed the flow separation by increasing the leading edge and improving the distribution of the inlet angle of attack. The load distribution of rotors with a 50% spanwise position changed from the original "C" loaded to post-loaded. The leading load obviously decreased, and the angle of attack was smaller than that of the prototype, which effectively weakened the flow separation at the leading edge of the rotor. Compared with the original rotor, the higher lean angle and pressure ratio of the turbine stage also improved. However, the leakage loss near the shroud of the rotor increased, which led to decreased efficiency.