MULTI-OBJECTIVE OPTIMIZATION IN AXIAL COMPRESSOR DESIGN USING A LINKED CFD-SOLVER

Kurzfassung

ABSTRACT In combination with a multi-objective 3D optimization strategy, a linked CFD-solver is introduced in this paper, combining 3D-Reynolds-averaged-Navier-Stokes and an inviscid through-flow method. It enables the adjustment of the 3D boundary con-ditions for any design variation and contains new options for configuring the objective functions. The linking is achieved by matching the flow information between both CFD codes in an iterative procedure. Compared to an individual 3D-CFD calculation, the convergence does not take significantly longer. The potential of the linked CFD-solver is demonstrated in a multi-objective optimization for one blade row to be optimized and one operating point at a 3-stage axial compressor with inlet guide vane. In the optimization, the objective functions are formulated, so that the performance of the axial compressor is enhanced in addition to the improved efficiency in the 3D-cascade.