IMPROVED TURBULENCE MODEL FOR AERODYNAMIC FLOWS WITH MASSIVESEPARATION AND WAKES

WE PROPOSE HERE A RATIONAL PROGRAM OF FUNDMENTAL RESEARCH TODEVELOP A MORE UNIVERS TURBULENCE MODEL FOR AERODYNAMIC FLOWS WITH MASSIVE SEPARATION AND WAKES. THIS PROGRAM IS BASED UPON THE OBSERVATION AND DEMONSTRATION BY BARDINA ET AL. THAT STRAIN AND ROTATIO HAVE DIFFERENT EFFECTS ON TURBULENCE AND TO ACCOUNT FOR THESE EFFECTS SEPARATELY IN MODELING CAN PRODUCE IMPROVED AGREEMENT WITH EXPERIMENT. THE PRESENT PROGRAM EXTENDS THE CITE WORK TO THE MORE GENERAL CASE OF SHEAR FLOWS WHERE NONLINEAR EFFECTS OF BOTH STRAIN AND ROTATION DETERMINE THE PRINCIPAL SCALES OF THE TURBULENCE. IT IS SHOWN IN THIS PROPOSAL THAT MOST TURBULENCE MODELS DO NOT ACCOUNT FOR THE EFFECTS OF ROTATION; THEREFORE, THESE MODELS CANNOT BE ACCURATE MODELS FOR THE PREDICTION OF SHEAR FLOWS, AND FAIL MORE NOTICEABLY IN FLOWS WHERE ROTATION IS PROMINENT SUCH AS IN SEPARATED AND WAKE FLOWS. IN PHASE I OF THIS PROGRAM, WE PROPOSE TO EXTEND THE MODEL OF BARDINA ET AL TO ACCURATELY ACCOUNT FOR THE EFFECTS DUE TO THE COMBINATION OF ROTATION AND TURBULENCE BY MEAN VELOCITY GRADIENT AND BY STRETCHING OF TURBULENT VORTICITY. THE MODEL WILL BE DEVELOPED BASED ON HIGHER-ORDER RAPID DISTORTION THEORY AND AVAILABLE INFORMATION PROVIDED BY FULL SIMULATIONS AT NASA-AMES RESEARCH CENTER, AND WILL BE TESTED AGAINST EXPERIMENTAL ANDFULL SIMULATION RESULTS FOR HOMOGENEOUS SHEAR FLOWS. IN PHASE II OF THIS EFFORT, WE WILL EXTEND THE MODEL TO INCLUDEDIFFUSION EFFECTS AND TEST THE MODEL FOR THE PREDICTION OF TURBULENT SEPARATED FLOWS AND WAKE FLOWS.