Three-dimensional subsonic diffuser design optimization and analysis

A novel methodology is developed to integrate state-of-the-art CFD analysis, the Non-uniform Rational B-Spline technique (NURBS) and optimization theory to reduce total pressure distortion and sustain or improve total pressure recovery within a curved three dimensional subsonic diffuser. Diffusing S-shaped ducts are representative of curved subsonic diffusers and are characterized by the S-shaped curvature of the duct's centerline and their increasing cross-sectional area. For aircraft inlet applications the measure of duct aerodynamic performance is the ability to decelerate the flow to the desired velocity while maintaining high total pressure recovery and flow near-uniformity. Reduced total pressure recovery lowers propulsion efficiency, whereas nonuniform flow conditions at the engine face lower engine stall and surge limits. Three degrees of freedom are employed as the number of independent design variables. The change of the surface shape is assumed to be Gaussian. The design variables are the location of the flow separation, the width and height of the Gaussian change. The General Aerodynamic Simulation Program (GASP) with the Baldwin-Lomax turbulence model is employed for the flow field prediction and proved to give good agreement with the experimental results for the baseline diffuser geometry. With the automatic change of the design variables, the configuration of the diffuser surface shape is able to be changed while keeping the entrance and exit of the diffuser unchanged in order to meet the specification of the engine and inlet. A trade study was performed which analyzed more than 10 configurations of the modified diffuser. Surface static pressure, surface flow visualization, and exit plane total pressure and transverse velocity data were acquired. The aerodynamic performance of each configuration was assessed by calculating total pressure recovery and spatial distortion elements. The automated design optimization is performed with a gradient-based method to minimize the total pressure distortion. Automated optimization processes have been performed based on two design variables. The best configuration obtained to date reduced distortion by 80 percent while keeping the total recovery roughly same. The results indicate that mechanism responsible for improved aerodynamic performance is the suppression of detrimental secondary flows by changing the surface shape to redirect the flow.