Multi-Disciplinary Optimization of a Centrifugal Compressor for Micro-Turbine Applications

The present paper presents the multi-disciplinary optimization of a centrifugal compressor for a 100kW micro gas turbine. The high rotational speed fixed by the cycle optimization (75,000 rpm) required a simultaneous analysis of flow aerodynamics and mechanical behavior to account for the high centrifugal stresses the blades are subjected to, while maximizing the aerodynamic performance. A commercial 3D (three dimensional) computational fluid dynamics (CFD) solver adopted for the aerodynamic computations and an open source finite element FEM code for the mechanical integrity calculations have been coupled with metamodels to speed up the optimization process. Home-made scripting modules, which manage multidisciplinary optimization, mesh generation, geometry parameterization and result post-processing have been written and utilized. A sample data-base has been generated on the basis of the parameters selected to describe aerodynamic and mechanical constraints, and an optimization procedure based on a genetic algorithm has been performed. A RANS (Reynold Averaged Navier Stokes) steady approach with a two-equation SST (Shear Stress Transport) model has been adopted for the aerodynamic computations during the optimization procedure. The optimized compressor so achieved showed an important boost in aerodynamic performance, without any penalty in the mechanical behavior, as compared with the preliminary design. The optimized configuration has been tested also by means of URANS (Unsteady Reynolds Averaged Navier Stokes) phase-lag investigations, which confirmed the aerodynamic performance increase predicted by steady RANS calculations.