This paper presents a design process coupling a 3D inverse design approach with a Multi Objective Genetic Algorithm (MOGA) for axial cooling fan. The aim is to reduce the leakage loss and profile losses to improve performance. The inverse design method parameterizes the 3D shape of the axial fan with a reduced number of design parameters allowing a larger exploration of the design space in the optimization process.
Two designs from the optimization are analyzed in detailed by using 3D Computational Fluid Dynamic (CFD) calculations, which confirm that the design optimized for minimizing leakage losses meets the design specification. Detailed CFD also confirms the reduction in leakage losses in this design.