Aerodynamic optimization of axial turbomachinery blades using parallel adaptive range differential evolution and Reynolds-averaged Navier-Stokes solutions

An automatic aerodynamic optimization, which coupled with parallel adaptive range differential evolution (PARDE), blade profile parameterization and Reynolds-averaged Navier-Stokes (RANS) solver, is developed for three-dimensional axial turbomachinery blades in this paper. RANS is utilized to evaluate the aerodynamic performance of blade design candidates. The non-uniform B-Spline method is used to parameterize the three-dimensional turbomachinery blade profile. PARDE is used as the optimizer and parallel algorithms using PC cluster technology are applied to accelerate the design period. Aerodynamic redesigns of the NASA rotor 37 and a turbine stage are utilized as optimization cases to demonstrate the reliability of the present optimization methodology. The rotor 37 is optimized for the maximization of the isentropic efficiency at the given flow condition, using constraints on the mass flow rate and the total pressure ratio. The isentropic efficiency of the optimum design rotor 37 has 1.66% higher than that of the reference design. A low aspect ratio transonic turbine stage is redesigned for the maximization of the isentropic efficiency with 41 design variables in total. The isentropic efficiency of the optimum turbine stage is 1.6% higher than that of the reference design. Copyright (C) 2009 John Wiley & Sons, Ltd.