High Reynolds Number CFD Benchmark: Introduction and Overview of Wind Tunnel Test Program

Abstract

Recent field measurements indicated that Vortex Induced Motion (VIM) of column-stabilized offshore platform was much smaller than predicted from scaled model test with Reynolds number two orders of magnitude lower than full scale. In order to understand the physical mechanisms that may have caused the discrepancy, Computational Fluid Dynamics (CFD) was applied in previous study to investigate the Reynolds number effect. It was shown that model-scale CFD simulation results agreed well with test data while full-scale CFD simulations indeed predicted a reduction in response amplitude. However, due to lack of well-controlled full-scale test data, it is hard to quantify the accuracy of full-scale CFD simulations. To fill this gap and provide benchmark data for CFD validation, we performed wind tunnel tests in a high pressure wind tunnel in DNW at Göttingen, Germany. The tests were done for single and tandem square column(s) with rounded corners and covered a wide range of Reynolds number from 4 x 10^5 to 1.2 x 10^7. Time history of total fluid forces on the column, time averaged surface pressure at two locations on column surface, and pressure profile in the middle wake were measured for each test. Surface oil-flow technique was applied in selected tests to visualize the flow and interpret the test results. The effects of variation in flow headings and column corner radius were also investigated. A selected subset of the test data are now provided for blind validation with several CFD software/practitioners. This presentation introduces the background of the study and gives an overview of the wind tunnel test program.