An optimum design procedure for an aerodynamic radial diffuser with incompressible flow at different Reynolds numbers

Abstract

In this article a systematic procedure aimed at achieving the best compromise between flow deflection, static pressure recovery, and total pressure loss is proposed for aerodynamic diffusers with incompressible flow. Such a result was accomplished by using a neural network to generalize the radial diffusers performance data obtained by numerical analyses, a multi-objective approach based on the employment of fuzzy sets, and a swarm particle algorithm to find a good compromise between flow deflection, static pressure recovery, and total pressure loss. Useful design tools, obtained by collecting the results in proper design charts, are finally proposed to simplify the design of radial diffusers without resorting to expensive and time-consuming procedures of optimization. The influence of the Reynolds number on the overall performance was also taken into account.