High performance computing at Sandia National Laboratories is playing a key role in the U.S. Department of Energy’s Wind and Water Power Technology Office mission of advancing the commercialization of tidal energy converters; by improving their power performance and reducing their levelized cost of energy below the local “hurdle” price at which they can compete with other regional generation sources without subsidies.
The Sandia-ORPC team applied high fidelity modeling to evaluate the performance of the RivGen® prototype turbine generation unit, a cross-flow turbine, which exhibits more complex flow physics than the more common axial-flow turbine. The 3D unsteady Reynolds-averaged Navier-Stokes (URANS) models used to predict power performance were first validated using a unique set of field measurements collected by ORPC in Cobscook Bay in 2014.
Numerical experiments, simulated on Glory, were then conducted to investigate and quantify parasitic drag effects on turbine performance and how these effects could be mitigated to improve performance. The results of this investigation provided a clear path for modifications to be made in the next design iteration of the RivGen turbine.
This study demonstrated the value of high fidelity modeling, and Sandia National Laboratories’ HPC resources, when resolving the complex 3D flow effects on performance that are sometimes encountered with complex turbine architectures.