Topology optimization using potential flow analysis

In current engineering fluid flow systems, space is restricted and traversing 90 degree turns or bends while trying to minimize pressure loss and maintain flow uniformity can be a challenge. Attempts to solve this problem have mainly been through the use of experimental and numerical trial and error methodologies to determine the shape, number, and arrangement of turning vanes, which invokes some experiential knowledge or intuition. Topology optimization presents a general design optimization approach that can produce non-conventional designs. Recently, the fluid dynamics community has adopted topology optimization for low to moderate Reynolds number (Re) flows, but research is lacking for moderate to higher Re, where most of these turning devices operate. As Re increases, the computational expense dramatically increases such that a low-fidelity model is needed to make topology optimization attractive. In this presentation, a topology optimization technique using a low-fidelity potential flow model for initial topology optimization is shown. A method for mapping the potential flow solution to physical boundaries that can be used in a viscous flow solver will be presented, followed by analysis of the topology using a high-fidelity viscous flow solver.