Multi-Objective Optimization for the Transonic Compressor Stator Blade

The two-dimensional transonic compressor stator was aerodynamically optimized as multi-objective problems (MOPs) using Multi-Objective Genetic Algorithms (MOGAs). For the objective function minimization of pressure loss coefficient and deviation outflow angle at a design point, and an incidence toughness to optimize more realistic conditions, were considered. The objective functions were calculated by the two-dimensional Navier-Stokes (NS) equations using a k-e turbulence model to evaluate the objectives with high accuracy. The calculation time, however, was large and parallel computing using Message Passing Interface (MPI) was adopted to decrease the total computation time. This made good use of the characteristics of GAs. Computation time was reduced by a factor equal to the number of CPUs used in the parallel computing. Though blade shapes for optimized results were different from those in general use, the optimized blades were found to have reasonable shapes for critical design conditions namely, the transonic flow regime and the high turning angle.