Comparison of Surface and Near Wake Flow Quantities Using URANS and LES Simulations
Abstract
The ability to dissipate heat produced by various components is vital toward the successful operation of various Naval platforms. Computational fluid dynamics offers a method to predict these vital flow quantities. This research characterizes both unsteady Reynolds-Averaged Navier-Stokes (URANS) and large eddy simulations (LES) to examine the advantages and limitations associated with each of these methodologies by comparing the predicted results with experimental data for flow past a cylinder and a mast. These simple geometries elicit complex flow fields posing a challenge for each methodology. URANS is best for attached turbulent flow with the ability to predict surface pressures correctly, even with relatively low grid resolutions. This makes it an effective tool if a quick turnaround is desired. Surface heat transfer requires properly resolving the temperature gradient and thermal boundary layer. As such higher resolution is required, LES is most effective in this regard, but requires significantly more grid resolution to resolve the small eddies. As such, its usefulness is still limited to the lower velocities. Local flow velocity profiles further highlighted the effectiveness of LES in predicting the correct flow velocities. Both methods have their uses for engineering level problems.
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 01, 2021
- Accession Number
- AD1121950
Entities
People
- Raymond Roberts
- Stephen A. Huyer
Organizations
- Naval Undersea Warfare Center