Optimization of pre-swirl nozzle shape and radial location to increase discharge coefficient and temperature drop

Abstract

Hole-type pre-swirl nozzle was optimized using CFD analysis and experiments. CFD methodologies were validated by comparing the CFD results with experiments. Four design variables were considered in the optimization process: Nozzle inlet length (L), outlet length (l), inlet diameter (D), and radial location (r(p)). The optimization process included the optimal Latin hypercube design sampling method with the Kriging surrogate model and genetic algorithm. The single-objective optimization was performed to maximize the discharge coefficient. Results showed that the optimized nozzle reduced total pressure losses and increased mass flow rate. Total temperature drop effectiveness was increased from 0.07 to 0.29. The total temperature in pre-swirl system could be characterized as the reduction in temperature by nozzle acceleration and elevation by aerodynamic losses due to friction and viscous effects in the system. The optimized model showed a discharge coefficient of 0.846, which was 31.7 % higher than the baseline condition. By improving the discharge coefficient, the pre-swirl system reduced aerodynamic losses, and the mass flow rate was increased at certain pressure ratios or satisfied the pressure margin for blade cooling.