Computational Fluid Dynamics
OverviewSTAR-CD is a mature high-performance computational fluid dynamics (CFD) environment that has been licensed for TRACC from CD-adapco. STAR-CD can perform reliable analysis of complex, multiscale transport phenomena in realistic industrial systems. STAR-CD is well suited to the solution of large-model simulations that benefit from efficient solution algorithms, memory utilization, and scalability on multiple processors. It features a well-integrated platform for creating high fidelity models from concept, or body-fitted meshes from existing CAD geometry models. It has robust solver technology for powerful multi-physics simulations involving turbulence, heat transfer, and reacting and multiphase flows. It also provides a means for users to add new physics that can be expressed as source terms in the differential equations via user defined subroutines.
The recent advancements in STAR-CD´s CAD-embedded technology and optimized polyhedral mesh generation dramatically improve its productivity and reliability, and enable analysts with widely varying skills and experience to gain insights from CFD
STAR-CD is also equipped with a rich selection of Reynolds-Averaged Navier-Stokes (RANS) turbulence models, versatile code-coupling capability, and powerful custom programming utilities, making it well suited to a wide variety of applications of interest to the TRACC community.
The STAR-CD license allows an unlimited number of concurrent jobs, and they can use all of the available cores.
Current TRACC ApplicationsTRACC, Turner-Fairbank Highway Research Center (TFHRC), and researchers at the University of Nebraska and Northern Illinois University are collaborating on the study of CFD-based simulation techniques. Researchers are taking reduced-scale experiments from the TFHRC hydraulics laboratory, providing the data for CFD model development, and producing a validated CFD-based advanced simulation methodology for open-channel flow, with an emphasis on bridge scouring.
The applicability of commercial CFD codes such as STAR-CD for prediction of these phenomena is being investigated, and the agreement between the code predictions and experimental data will be determined for various modeling options. The scalability of these simulations to large numbers of processors, particularly for the simulation of full-scale bridge deck interactions, is being evaluated and guidelines will be developed for the decomposition of problems of this type. Cross-code comparisons of the calculated results to evaluate computational efficiency and accuracy are also under investigation.