PARALLEL MULTILEVEL ADAPTIVE METHODS FOR FLOWS IN TRANSITION

THE PURPOSE OF THIS PROJECT IS TO DEVELOP AN ACCURATE AND HIGHLY EFFICIENT PARALLEL ALGORITHM BASED ON ASYNCHRONOUS, MULTILEVEL, ADAPTIVE METHODS FOR COMPUTATIONAL FLUID DYNAMICPROBLEMS INVOLVING TRANSITION CAUSED BY CANONICAL ROUGHNESS ELEMENTS. BOUNDARY LAYER TRANSITION XTREMELY SENSITIVE TO THE PRESENCE AND NATURE OF ROUGHNESS ELEMENTS, BUT VERY LITTLE IS KNOWN EITHER EXPERIMENTALLY OR NUMERICALLY ABOUT THE PRECISE NATURE OF THIS RELATIONSHIP. THE ULTIMATE GOAL OF THIS PROJECT IS TO DEVELOP EFFECTIVE COMPUTATIONAL TOOLS FOR NUMERICAL ANALYSIS OF THESE PHYSICAL PROCESSES. THE PHASE I PROJECT WILL DEVELOP A PROTOTYPE METHOD FOR TWO- DIMENSIONAL AND THREE-DIMENSIONAL, STEADY, INCOMPRESSIBLE FLOW IN A CHANNEL. THE APPROACH WILL INVOLVE THE DEVELOPMENT OF THE ASYNCHRONOUS, FAST, ADAPTIVE, COMPOSITE- GRID METHOD FOR SUCH PROBLEMS ON THE INTEL IPSC/2 USING SEVERAL DIFFERENT ROUGHNESS ELEMENTS, HEMISPHERE, SPHERE, AND CYLINDER. THE PROJECT SHOULD ENABLE PARALLEL COMPUTER SIMULATION OF STEADY-STATE AND TIME-DEPENDENT SIMULATION OF THREE- DIMENSIONAL FLOWS IN TRANSITION INVOLVING CANONICAL ROUGHNESS ELEMENTS.