A Brief Investigation of the Repeatability and Reproducibility Of PIV Measurements for CFD-grade Databases

摘要

Over recent decades, Particle Image Velocimetry (PIV) has become an increasingly common optical technique used to study fluid flows. As a result, large quantities of PIV data are becoming readily available for the development and validation of Computational Fluid Dynamics (CFD) codes. The CFD community heavily relies on experimental data, and the experimental community should focus on an important quality that is often overlooked - the repeatability and the reproducibility of PIV data. Unfortunately, with advanced instrumentation such as a PIV system, slight variations in equipment setup can be sources of discrepancies in the measurements. In the present work, we discuss the level of repeatability and reproducibility of PIV results obtained in the Reactor Cavity Cooling System (RCCS) separate-effects test facility - a facility that aims to provide a high-resolution experimental database to be used to further develop the predictive capabilities of system and CFD codes. We examine the repeatability and reproducibility of the mean velocity and Reynolds stresses for the experiment of six rectangular jets with jet Reynolds numbers of 1.38×10~4. Furthermore, we discuss the convergence of the mean velocity and Reynolds stresses and some discrepancies observed between separate measurements. By conducting a few PIV measurements to verify the repeatability and reproducibility of the data, we gained insight into how much variation is present between separate measurements - an important quantity that should be included when providing uncertainty bands for PIV data.